UNIVERSITY OF OKLAHOMA

GRADUATE COLLEGE

PARODIES AND THE ROLE OF MEDIEVAL ETHICAL CONCEPTS IN THE

ONTOLOGICAL ARGUMENT

A DISSERTATION

SUBMITTED TO THE GRADUATE FACULTY

in partial fulfillment of the requirements for the

Degree of

DOCTOR OF PHILOSOPHY

By

SAMMUEL ROBERT BYER Norman, Oklahoma 2017

PARODIES AND THE ROLE OF MEDIEVAL ETHICAL CONCEPTS IN THE ONTOLOGICAL ARGUMENT

A DISSERTATION APPROVED FOR THE DEPARTMENT OF PHILOSOPHY

BY

______Dr. Linda Zagzebski, Chair

______Dr. Neal Judisch, Co-Chair

______Dr. James Hawthorne

______Dr. Edward Sankowski

______Dr. Mara Willard

© Copyright by SAMMUEL ROBERT BYER 2017 All Rights Reserved.

For the Divine and for my family: Samuel Byer, Anne Byer, and Sally Spina. Acknowledgements This dissertation and my own philosophical development have been the product of many different people. Though any list such as this will be incomplete, I wish to acknowledge my debt and gratitude to the following people, who shaped and formed my thought in a number of ways. My parents, Sam and Anne Byer; my sister and brother-in-law, Sally Spina and Dan Spina; my friends, Brandon Rouse, Jacob Cochran,

Ryan Leiker, Paul Arias, Mohsen Mahmoud, Jaremia Green, Qing Guo, Andrew Russo,

Zach Jurgensen, Steve Dwyer, Jessica Jones, Julie Dwyer, and Casey Stull; and my teachers, Stephen Tramel, Gene Rice, Rich Hughen, Doug Drabkin, Steve Trout, Brad

Will, Tom Boyd, Barbara Boyd, Erik Braun, Hugh Benson, Mara Willard, Ray

Elugardo, Amy Olberding, Ed Sankowski, Chris Swoyer, and Jim Hawthorne. Finally,

I wish to acknowledge the great personal and philosophical debt I owe to both Neal

Judisch and Linda Zagzebski.

iv

Table of Contents

Acknowledgements……………………………………………………………………..iv

Abstract…………………………………………………………………………………vi

Chapter 1: Ontological Arguments and the Structure of the Project…………………….1

Chapter 2: Senses of Existence…………………………………………………………23

Chapter 3: Parody Objections and Anti-God…………………………………………...49

Chapter 4: Being and Goodness………………………………………………………..93

Chapter 5: The Central Argument…………………………………………………….137

References………………………………………………………………………….…188

v

Abstract

In this dissertation, I examine historical and contemporary versions of the ontological argument for God’s existence and objections to it, focusing on versions of the argument that contain Linkage Premises (premises that hold there is a strong connection between existence and perfection or greatness). I argue that the most plausible kind of objection to these and other ontological arguments is a version of the parody-style counterargument found in contemporary literature. If this counterargument succeeds, it results in a reductio ad absurdum of the ontological argument. I then examine the relationship between being and goodness found in certain medieval philosophers such as Aquinas, Augustine, and Boethius (also called the

Convertibility Thesis of Being and Goodness). I argue that this view is coherent and plausible. Then, I argue that for versions of the ontological argument that contain a

Linkage Premise to succeed, both against my counterargument and on their own terms, the proponent must presuppose the Convertibility Thesis of Being and Goodness.

Finally, I argue that greater attention ought to be paid to this medieval view, not only for its positive results related to the ontological argument, but also because of its relevance in other ethical, metaethical, and metaphysical matters.

vi

Chapter 1: Ontological Arguments and the Structure of the Project

In this chapter, I will examine several ontological arguments, both recent and historical. With the exception of Anselm’s argument in Proslogion II, all are modal ontological arguments. I will also explain the purpose of the dissertation as a whole by giving a precis of each chapter and its relation to the other chapters of the dissertation.

Section I. Introduction

The ontological argument occupies a strange and interesting role in the history of philosophy. First proposed by Anselm of Canterbury, the ontological argument moves from considerations regarding the concept or definition of God and the kinds of properties or essential features required for Godhood to an argument that gives us reason to think that God exists. Throughout modern Western intellectual history, a number of important philosophers have proposed their own ontological arguments for the existence of God, and just as many philosophers have taken the argument to task.

Descartes, Leibniz, Kant, and Hume all had something to say about the ontological argument, whether they thought it to be successful or unsuccessful. The development of new modal logics resulted in a return of the ontological argument in 20th century analytic philosophy. Taking Anselm as giving a central insight into the nature of God’s existence as necessary, a number of 20th century philosophers attempted to refine and further develop the ontological argument with the newly developed resources at their disposal. Even now, ontological arguments and various objections to those arguments figure prominently in contemporary journal articles and scholarly monographs.

All of this is sufficient to demonstrate that the ontological argument has a well- worn history, but I find it to be philosophically important for one other major reason. In

1 the ontological argument there are a number of important metaphysical, modal, and moral concepts at play, and any attempt to work through the argument will require grappling with a host of fundamental philosophical positions. What does it mean for something to exist? Can there be anything that necessarily exists? What is it for something to be possible? Do our own thought processes and patterns of reasoning give us any justification for modal knowledge? What does it mean for something to be good? How, if at all, are greatness and goodness related? These are just some of the questions, utilizing concepts, found at the heart of traditional and contemporary ontological arguments. The ontological argument serves an important role for general metaphysical issues regarding theism and atheism as well. In addition to being a deductive argument, the ontological argument if it is valid and sound can serve to undercut or undermine the problem of evil and all other atheological arguments. In fact, if the argument works it shows that there are no possible worlds where God fails to exist. All of this shows that the argument, stretching backwards and history and forward into the future, is truly worth considering.

Section II. The Proposal

As a whole, this dissertation aims to make three primary arguments. First, that a central and common objection to the ontological argument is actually significantly weaker than many philosophers have thought. Second, that a style of objection often taken to be the least viable and least troublesome is actually the most powerful objection to the ontological argument. Finally, that the ontological arguments I’ll discuss here require the convertibility thesis of being and goodness to be true for their success. In some cases this requirement is explicit and conceptual, whereas in other

2 cases I will argue that the defenders of the ontological argument are at least practically committed to the truth of the convertibility thesis of being and goodness. The dissertation progresses through five chapters.

The first chapter contains a standard form preparation of each of the ontological arguments that I focus on in later chapters, especially chapter five. I will not try to explicate these arguments in any great detail in this chapter, since the most relevant sections of these arguments receive much greater scrutiny in the final chapter. This chapter also contains a preview of the dissertation.

The second chapter explores different views of what it means to say that an object exists. The current orthodox view and its historical predecessors are examined in some detail. I also examine alternative concepts of existence. The purpose of this chapter, like the next, is in part conceptual mapping. I desire to know what we need to assume or commit ourselves to for the ontological argument to be a viable, sound argument, at least as regards the nature of existence. I finally conclude that 1) ontological arguments are upheld if existence is a first-order property, since there are alternative accounts of the existence predicate and whether or not it applies to objects, and that 2) ontological arguments are upheld if existence is not a first-order property, in the style of Malcolm and Plantinga. Any way the objection is lodged, this most famous and global critique of the argument has little to recommend it – ontological arguments may proceed regardless of the nature of existence.

The third chapter takes up the idea of developing a parody style objection to ontological arguments. In this chapter, I present a modified taxonomy of parody style objections, and I give an argument that parody style objections are the most powerful

3 objections against ontological arguments. I also give an argument for preferring one particular style of parody objection over others, which I call the AntiGod objection. I give an example of this kind of objection and consider a rebuttal that all parody style arguments must fail. At the end of the chapter, if I have succeeded I will have shown that AntiGod parody objections are not only not destined to fail, but are the most powerful stumbling block to ontological arguments – particularly in comparison to the two previous objections discussed in chapters two and three.

The fourth chapter focuses on developing the convertibility thesis of being and goodness. In this chapter I briefly examine the historical pedigree of the doctrine before giving an in-depth analysis of what I take to be the best contemporary form of this thesis. In explaining the convertibility thesis of being and goodness, I take care to defend it against the objections that it is no longer suitable as a metaphysical/metaethical theory of our reality. I show that the convertibility thesis of being and goodness can be used to analyze goodness in a variety of forms. I also show that the convertibility thesis of being and goodness, in at least one form, does not require God’s existence – that a philosopher could develop this view without being a theist. This goes part of the way to show that the convertibility thesis can still be effectively and faithfully utilized in contemporary philosophy. In the course of defending the convertibility thesis, it will not be enough to simply show that we can make sense of the contemporary world by adopting that lens. The single most persuasive objection to the convertibility thesis of being and goodness is that it entails the privation theory of evil, that all evil can be analyzed as a privation, or absence, or lack. And the single most powerful counterexample to the privation theory of evil is the

4 example of pain. In order to defend the convertibility thesis, in this chapter I focus on defanging the objection from pain, and show that pain does not constitute a sufficient obstacle to rule out the privation theory of evil, and in turn, the convertibility thesis of being and goodness.

The fifth and final chapter gives my own, multi-stage argument that each version of the ontological argument covered in this chapter is committed to the truth of the convertibility thesis of being and goodness. In that chapter I demonstrate the grounds for thinking that the ontological arguers in fact require this thesis, as well as giving an argument that the convertibility thesis of being and goodness is superior to any alternative theory of greatness or greatmaking properties on offer. The idea that this is required as a background for most ontological arguments is either completely accepted without mention in contemporary discussions of the ontological argument, its importance has been widely overlooked if accepted, or most people have not noticed this until now. If my argument is correct, it will no longer be possible to develop ontological arguments of a certain type without taking into account the need for the convertibility thesis of being and goodness. The convertibility thesis is not simply superior to the other great-making theories offered by ontological arguers, but for wider ontological application. As this is the final chapter, I also summarize the results of the dissertation.

Section III. Notes on Formulating the Arguments

In the following sections I give versions of ontological arguments by Anselm of

Canterbury, Rene Descartes, Alvin Plantinga, Kurt Gӧdel, Norman Malcolm, Robert

Maydole, and Gottfried Wilhelm Leibniz. I examine these arguments because I believe

5 that, although this list is not exhaustive, it is a good cross-section of ontological arguments that have occurred at various times throughout history. If my main argument in chapter five is correct, then the charge I level there will apply to each of these arguments. If it applies to each of these arguments, then this opens up the possibility that it will also apply to other kinds of ontological arguments. Concerning myself with these arguments gives the project a substantive base from which to draw further conclusions.

My primary focus in this chapter is not to lay out exact, nuanced versions of each argument listed. For any of the arguments given, numerous philosophers have given their own interpretation of the premises, or the text, or the logical structure of the argument, or even the conclusion, and they all argue for the supremacy of their own version. This is far beyond what is realistic for my own project. Instead, I try to formulate a version of the argument that does not do significant damage to the text it is drawn from, and that captures a basic but correct version what is going on in each argument. If the versions of arguments given in this chapter are basically accurate, then even if there is room to debate finer, nuanced points of interpretation or precision, it is unlikely to have any effect on my final argument. The main work of diving into each text to rigorously support my final argument will occur in the fifth chapter.

Section IV. Anselm’s Argument

Anselm’s argument is the first ontological argument, and it is probably the simplest to state. In what follows I’ll present my own interpretation of Anselm’s argument in Proslogion II and III.

Anselm’s argument in Proslogion II is given in short fashion as follows.

6

Therefore, 0 Lord, You who give understanding to faith, grant me to understand—to the degree You know to be advantageous—that You exist, as we believe, and that You are what we believe [You to be]. Indeed, we believe You to be something than which nothing greater can be thought. Or is there, then, no such nature [as You], for the Fool has said in his heart that God does not exist? But surely when this very same Fool hears my words “something than which nothing greater can be thought,” he understands what he hears. And what he understands is in his understanding, even if he does not understand [i.e., judge] it to exist. For that a thing is in the understanding is distinct from understanding that [this] thing exists. For example, when a painter envisions what he is about to paint: he indeed has in his understanding that which he has not yet made, but he does not yet understand that it exists. But after he has painted [it]: he has in his understanding that which he has made, and he understands that it exists. So even the Fool is convinced that something than which nothing greater can be thought is at least in his understanding; for when he hears of this [being], he understands [what he hears], and whatever is understood is in the understanding. But surely that than which a greater cannot be thought cannot be only in the understanding. For if it were only in the understanding, it could be thought to exist also in reality—something which is greater [than existing only in the understanding]. Therefore, if that than which a greater cannot be thought were only in the understanding, then that than which a greater cannot be thought would be that than which a greater can be thought! But surely this [conclusion] is impossible. Hence, without doubt, something than which a greater cannot be thought exists both in the understanding and in reality. (Anselm 2000a, pp. 8-9, bracketed phrases from the translator)

The general strategy most reflective of Anselm’s argument here is the reductio ad absurdum. In standard form, the argument would proceed as follows.

1. God is defined as “the being than which nothing greater can be conceived.” (Definition) 2. God exists in the understanding but not in reality. (Atheist’s Assumption) 3. It is greater to exist in reality and the understanding than it is to exist in the understanding alone. (Anselm’s Principle) 4. It is possible to conceive of a being greater than the God in premise 2, namely a God that exists in both understanding and reality. (2, 3) 5. If 4, it is possible to conceive of something greater than the being than which nothing greater can be conceived. (Contradiction, 1, 4) 6. Therefore, God exists in the understanding and in reality.

7

There are numerous ways one might approach or attack this argument. A great degree of scholarly work has been spent trying to figure out how to correctly and accurately map out this argument. For our purposes, as long as my translation does not do some considerable injustice to Anselm, this will suffice.

It is fairly easy to see how the argument goes. One begins with the Atheist’s

Assumption for the purpose of reductio. Armed with this assumption, the definition of

God as the greatest conceivable being, and the principle that existing in reality and the understanding is greater than existing in the understanding alone, we can easily generate a contradiction. As Anselm says, we accept 2 and 3, we can show that there could be something greater than the Atheist’s God, a God which exists both in reality and in the understanding. But if that is true, it looks like we can conceive of something greater than God – something than which nothing greater can be conceived. Since this is contradictory, we must reject one of the previous assumptions. The definition of God seems rather straightforward, and for reasons I will present later Anselm has good reason to hold that the principle linking greatness to existence is true. If all of this is right, then we must reject the Atheist’s Assumption. If we do, we have left two options

– that God does not exist in the understanding or reality, or that God exists in both reality and in the understanding. The first is false since even for the Atheist, God exists in the understanding. Therefore, God must exist in both the understanding and reality.

Therefore, God exists.

The argument given in Proslogion 3 is quite a bit different. I quote the full section below.

Assuredly, this [being] exists so truly [i.e., really] that it cannot even be thought not to exist. For there can be thought to exist something which

8

cannot be thought not to exist; and this thing is greater than that which can be thought not to exist. Therefore, if that than which a greater cannot be thought could be thought not to exist, then that than which a greater cannot be thought would not be that than which a greater cannot be thought—[a consequence] which is contradictory. Hence, something than which a greater cannot be thought exists so truly that it cannot even be thought not to exist. And You are this [being], 0 Lord our God. Therefore, 0 Lord my God, You exist so truly that You cannot even be thought not to exist. And this is rightly the case. For if any mind could think of something better than You, the creature would rise above the Creator and would sit in judgment over the Creator—something which is utterly absurd. Indeed, except for You alone, whatever else exists can be thought not to exist. Therefore, You alone exist most truly of all and thus most greatly of all; for whatever else exists does not exist as truly [as do You] and thus exists less greatly [than do You]. Since, then, it is so readily clear to a rational mind that You exist most greatly of all, why did the Fool say in his heart that God does not exist? —why [indeed] except because [he is] foolish and a fool! (Anselm 2000a pp. 9)

This is the argument that Hartshorne and Malcolm rediscovered as a modal ontological argument. Below is my own interpretation of the argument in standard form.

1. God is defined as “the being than which nothing greater can be conceived.” (Definition) 2. It is possible to conceive of a necessarily existent being (a being that cannot be thought to not exist). (Assumption) 3. Necessary existence (beings that are unable to be thought to not exist) is greater than contingent existence (beings that are able to be thought to not exist). (Modified Anselm’s Principle) 4. God exists only contingently and not necessarily (assumption for reductio). 5. It is possible to conceive of a being greater than the God in 4, namely a God that exists necessarily. (3, 4) 6. If 5, it is possible to conceive of something greater than the being than which nothing greater can be conceived. (Contradiction, 1, 5) 7. Therefore, God necessarily exists.

As before, this argument starts out with the definition of God as the being greater than which none can be conceived. It then moves quickly into consideration of whether or not it is possible to conceive of a being that is necessarily existent. In

9

Anselm’s terms this is a being that “cannot be thought to not exist” but I think this captures our own notion of a necessarily existent being quite well. In another (for

Anselm) noncontroversial move, he adopts a modified version of the previous principle that states that this kind of existence (where an entity cannot even be thought to not exist) is better than the other kind (where an entity can be thought to not exist). We again have an assumption for reductio, which is that God might be thought to only contingently exist. Just as in the last argument, this leads us to a contradiction. Anselm gives up the idea that God might only contingently exist instead of the other premises.

If this is correct, then God not only exists, but exists necessarily.

Section V. Descartes’ Argument

Descartes has several arguments for the existence of God in the Meditations, but his version of the ontological argument occurs in the Fifth Meditation and the

Principles of Philosophy, and is further fleshed out in the Replies and Objections and personal correspondence (Descartes 2008). I will reproduce what I take to be one version of the argument here, taken from the Fifth Meditation.

1. I have the idea of God in my mind. (Assumption) 2. God is defined as the supremely perfect being. (Definition) 3. A supremely perfect being must possess all perfections in all ways. (Definition) 4. God cannot lack any perfection. (2, 3) 5. Whatever I clearly and distinctly perceive is true. (Clarity and Distinctness Principle) 6. I clearly and distinctly perceive necessary existence to be one among God’s perfections. (Assumption) 7. Necessary existence is one of God’s perfections. (5, 6) 8. Therefore, God necessarily exists.

There are undoubtedly other ways to give the form of this argument – some presentations focus on the essence-existence or attribute-concept distinctions in

10

Descartes’ argument, whereas others might forego the role of clear and distinct perceptions in coming to know that necessary existence is a perfection. Since this work is not primarily about historical scholarship, and since the argument given above does not unduly distort or damage Descartes’ own ontological argument, we can be comfortable in proceeding.

Like Anselm’s, Descartes version of the ontological argument flows directly and smoothly to its conclusion. Premise 1 is just the definition of God as the supremely perfect being, not dissimilar to Anselm’s own phrase “the being than which none greater can be conceived.” This sets the stage for the remaining arguments. Premise 2 seems to be an implication of 1, that a being supremely perfect would not lack any perfection in any way (on the pain of it not then being supremely perfect). Premise 3 just substitutes the proper name “God” for the phrase “a supremely perfect being” and rephrases “must have all possible perfections” to “cannot lack any perfection.” Premise

4 states that necessary existence is a perfection. Descartes grounds this by referring to clear and distinct perceptions, but he also spends a significant amount of time in the

Meditations developing what is very similar to the convertibility thesis of being and goodness in order to further support this point. If this argument is correct, God necessarily exists.

Section VI. Plantinga's Argument

I will present the argument that Plantinga gives in The Nature of Necessity

(Plantinga 1974). I will add two implications I take his argument to have, but that do not appear in the original form. I will use these terms just as Plantinga does; for example, “W*” denotes some possible world, and “E*” denotes some essence.

11

1. The property has maximal greatness entails the property has maximal excellence in every possible world. 2. Maximal excellence entails omniscience, omnipotence, and moral perfection. 3. Maximal greatness is possibly exemplified. 4. There is a world W* and an essence E* such that E* is exemplified in W* and E* entails has maximal greatness in W*. 5. For any object x, if x exemplifies E*, then x exemplifies the property has maximal excellence in every possible world 6. E* entails the property has maximal excellence in every possible world. 7. If W* had been actual, it would have been impossible that E* fail to be exemplified. 8. What is impossible and what is necessary do not vary from world to world. 9. There exists a being that has maximal excellence in every world. 10. So God exists. (Plantinga 1974, 214 – 215)

Premise 1, 2, and 3 give us the initial conditions of the argument. 1 and 2 are simply definitions Plantinga gives of the relevant terms “maximal greatness” and

“maximal excellence.” 3 is just the possibility premise. It merely states that it is possible that maximal greatness is exemplified or instantiated. Premise 4 is an implication of Plantinga's view about the exemplification of properties; for some property P to be possibly exemplified, is just for there to be an essence E in a world W where essence E entails has P in W. So 4 is just a more precise statement of 3. In premise 5, Plantinga suggests that if W* had been the actual world, then 5 would be necessarily true. However, what is necessary does not change from world to world, and

5 is necessary. If 5 is necessary, then we can restate what essence E* entails in 6.

According to Plantinga's view of how properties are exemplified, for a being to possess a property in a world W requires that it exist in world W. If E* entails “has maximal excellence in every possible world,” then it must also entail “exists in every possible world.” But if E* is exemplified in W*, then it must be exemplified by some entity or

12 other in W* and in all other worlds. All of this implies 7. Again, since what is necessary or impossible do not change from world to world, from premise 8, we find that it is impossible that E* fail to be exemplified in any world. But if that is right, we get 9. Of course, an implication of 9 is that there exists a maximally excellent being in our world, the actual world. So, God exists.

Section VII. Gödel's Argument

I will follow Graham Oppy's presentation of the argument (in Oppy 1995). I will also follow Oppy in taking Anderson's emendations to Gödel's argument as the best way to present it (in Anderson 1990). Sobel has shown that Gödel's initial formulation of the argument implies modal collapse – that is, every truth becomes a necessary truth, and all existent objects become necessarily existent (in Sobel 1987 and Sobel 2003).

With Anderson's slight alterations in the formulations of the definitions of “essence,”

“God-likeness” and Axiom 1, we get the following argument that does not imply modal collapse.

Definition 1. x is God-like iff x has as essential properties those and only those properties which are positive. Definition 2. A is an essence of x iff for every property B, x has B necessarily iff A entails B. Definition 3. x necessarily exists iff every essence of x is necessarily exemplified.

Axiom 1. If a property is positive, then its negation is not positive. Axiom 2. Any property entailed (= strictly implied by) by a positive property is positive. Axiom 3. The property of being God-like is positive. Axiom 4. If a property is positive, then it is necessarily positive. Axiom 5. Necessary existence is positive.

Theorem 1. If a property is positive, then it is consistent (=possibly exemplified). Corollary 1. The property of being God-like is consistent.

13

Theorem 2. If something is God-like, then the property of being God- like is an essence of that thing. Theorem 3. Necessarily, the property of being God-like is exemplified. (Oppy 2016)

While it is beyond the scope of this project to attempt to fully explicate or justify

Gödel's axioms, I will make some effort to explain what I take to be going on at each point, to aid the reader in understanding this version of the proof. I take more time with this proof than most of the others since I think it is the least clear and the most in need of unpacking.

Definition 1 is Gödel's definition of what it is to be God-like. If something is to be God-like, then Definition 1 tells us that that entity must essentially exemplify all and only those properties judged to be positive in the appropriate sense. Definition 2 is just the Anderson-altered, Gödelian view of essences and their relationship to different kinds of properties. If there are properties that are accidental as well as essential, then an acceptable account of essence will give us this difference. Definition 2 states that something is an essence for some entity, if and only if, for every property, the entity necessarily exemplifies that property if and only if the essence entails that property.

Definition 3 is an account of necessary existence; what it means for something to exist necessarily is just for every essence of that thing to be exemplified. Gӧdel’s definition of necessary existence here is stronger than the normal conception of necessary existence, in part because his concept of an essence is different than in other cases. An essence, on Gӧdel’s view, is some property that entails or necessitates each of the actual properties of the object (Koons 2005, Look 2006). Any object that existed necessarily would have all and only its essential properties necessarily, and none of them contingently. However, we usually think that there is a distinction between the

14 accidental properties of an object and the essential properties of an object. This is not true for Gӧdel’s concept of an essence, and so his version of necessary existence is also stronger, as it requires that for the essence of an object to necessarily exist will mean that the object has all and only necessary properties – none of its properties could be contingent.

Axiom 1 states a certain view of the relationships between positive properties, and their negations. Gödel states in his notes that a positive property can be positive in either the “moral-aesthetic sense” or in a sense of “pure attribution.” Is this true? Let us take omniscience as a test case. If omniscience is a positive property, then Axiom 1 tells us that its negation is not positive, again in the “moral-aesthetic sense” or “pure attribution” sense. Is this true? A lack of omniscience would not be positive in either of these senses. Thus, the property of “not being omniscient” or “lacking omniscience” is not itself a positive property. But since this is what Axiom 1 says, it is not a surprising result. This is perhaps the strangest part of the argument as a whole. Since there is so little to go off of in interpreting this argument, it is hard to know what counts as a positive property for Gӧdel, and exactly why he thought the two senses (moral- aesthetic and pure attribution) were connected. It is also difficult to know how to interpret “moral-aesthetic” and “pure attribution” as Gӧdel’s notes give no additional information to work with. The “moral-aesthetic” seems to be the one most of us are familiar with when we talk about goodness in terms of the character of persons or objects of art. The sense of “pure attribution” sounds similar to an account where positivity is some substantive, existence quality, and a negative property is not itself a separate and opposite quality but the absence of some positive property. This is a vague

15 and preliminary sketch, but Gӧdel’s notes tell us little. I will treat Gӧdel as if he is carrying on a project in the spirit of Leibniz, and so we can see positive properties as analogous to perfections in Leibniz. Several commentators have taken Gӧdel to be carrying on a project in the style of Leibniz, so this interpretation is standard (Koons

2005, Look 2006, and Sobel 2003).

Axiom 2 states that any property entailed by a positive property is itself positive.

Axiom 3 simply states that the property of being God-like is a positive property.

Surely, this is unobjectionable even to the most strident atheist. Axiom 4 is an assertion about the modal status of positivity in properties. If a property is positive, this is not the case contingently; it won't turn out that a property is positive, but might not have been so, or that a property could be positive in one possible world and not in another. If one is a robust realist about morality, then this axiom seems fairly intuitive. If one were to hold a different view about morality, then this axiom may be less convincing. Anderson also points out that what the proof really requires is the necessitation of this axiom.

That is, “that a property is positive entails that it is necessarily positive.” He also notes, though, that “probably no one who accepts the axioms will shrink from asserting all their necessitations,” (Anderson 1990, footnote 4).

Axiom 5 is, in at least some way, one part of the argument opponents will reject.

It states that necessary existence is a positive property. From Definition 1, we know that for something to be God-like, it must essentially exemplify all and only positive properties. If necessary existence is a positive property, then God will turn out to be (at the end point of the argument) necessarily existent. Which will mean that God exists, or better, that something that is God-like exists.

16

Theorem 1 is generated by Axioms 1 and 2. Suppose some property P is positive. If it is positive, then it entails the property of self-identity (since every property entails self-identity). Then the property of self-identity is positive, by Axiom

2. If that is so, then the negation of self-identity, self-difference, is not positive by

Axiom 1. But if P were inconsistent, it would entail the property of self-difference, since in classical logic an inconsistent property entails any property. But this is impossible, from Axiom 2. So every positive property is consistent. Theorem 1 combined with Axiom 3 gives us Corollary 1 – the property of being God-like is consistent. This provides a unique way of resolving the issue over what I take is the major premise of any ontological argument – whether or not the concept of God is a consistent, coherent concept.

Theorem 2 is generated by Definitions 1 and 2, and Axioms and 4. Suppose x is

God-like, and necessarily has property P. Property P must be positive, from Definition

1. If P is positive, then necessarily anything that is God-like will have that property, from Definition 1. From Axiom 4, we know that P is necessarily positive. If this is true, then the property of God-likeness essentially includes P. We can now substitute property P and God-likeness for the general placeholders in Definition 2. So, we get

Theorem 2 — if something is God-like, then the property of being God-like is an essence of that thing. Theorem 3 is proven by Definitions 1 and 3, Axiom 5, Corollary

1, and Theorem 2. Some x is a God-like being, then it will have every positive property, from Definition 1. From Axiom 5, necessary existence is positive, so from

Definition 1 and Axiom 5, necessary existence will be a positive property a God-like being has. Definition 3 states that x necessarily exists only if every essence of x is

17 necessarily exemplified. Theorem 2 states that for any God-like being, God-likeness is an essence of that being. So God-likeness is necessarily exemplified, from Definition 3 and Theorem 2. If it is possible that something is God-like, then it is possible that something is necessarily God-like. But Corollary 1 shows us that it is possible that something is God-like. So Theorem 3 is shown to be true. Something God-like necessarily exists.

Section VIII. Malcolm’s Argument

Norman Malcolm’s contribution to the recent literature on the ontological argument came in two forms. First, he was interested in showing that the actual arguments of Anselm were more complex and nuanced than most philosophers had taken them to be. Second, in “Anselm’s Ontological Arguments” he sees himself as

“trying to expand the argument of Proslogion III,” (Malcolm 1960, pp. 48). This suggests that in addition to historical scholarship, he is proposing his own formulation of an ontological argument.

1. God is defined as the greatest of all beings. (Definition) 2. The greatest of all beings must be unlimited – were it limited, it would not be the greatest of all beings. (Definition of greatest and unlimited) 3. An unlimited being could not depend on anything for its continued existence or its coming into existence – if it did so, it would be limited. (Definition of limited vs. unlimited) 4. Either God’s existence is necessary or God is necessarily impossible. (Implication of 1, 3) 5. There is no reason to think that the concept of God is internally contradictory or otherwise necessarily impossible. (Assumption) 6. The concept of God is possible. (From 5) 7. God necessarily exists. (from 4, 5, 6)

Implicit in premise 1 through 3 is an idea about the relationship between necessary existence and greatness. He gives an example of two sets of dishes, one

18 being fragile and the other being necessarily existent is supposed to generate a value judgment in the reader of one set of dishes being greater than the other (Malcolm 1960, pp. 47). We would make the same judgment about God – anything that plays that role could not be dependent and thus inferior, or it would not be the greatest of all beings.

Malcolm also draws on God’s being unlimited in order to support the idea of a necessary connection between the concept of God and necessary existence. For something to exist contingently suggests, according to Malcolm, some limitedness in the entity or object.

Malcolm draws on this insight to argue that whatever the fact of the matter about

God’s existence, the very nature of the being would not allow a contingent mode of existence (an argument previously developed by Hartshorne). God could not depend on something else for His continued existence or its end, nor could God’s existing in the first place be a contingent matter. On Malcolm’s view, it is all or nothing – either the concept of God is possible and God necessarily exists, or the concept of God fails to exist in every world due to its logically contradictory nature. Since no one has successfully demonstrated that the concept of God is internally logically contradictory, we are well within our epistemic rights to assert God’s possibility. Therefore, God exists.

Section IX. Maydole’s Argument

Maydole’s argument draws on considerations familiar to (and previously explored by) Anselm, Descartes, Leibniz, and Gӧdel. His argument in its simplest form proceeds as follows.

1. A property is a perfection only if its negation is not a perfection. 2. Perfections entail only perfections.

19

3. Supremity is a perfection. 4. A supreme being is a being that it is impossible for there to be something greater than and impossible for there to be something else than which it is not greater. 5. It is possible that a supreme being exists. 6. There exists a being who is possibly supreme. 7. Therefore, a supreme being (God) exists.

This is a simplification, as in different works Maydole proves the argument to be valid in 32 steps (Maydole 2003) and 16 steps (Maydole 2009). It is similar to both

Descartes and Anselm’s arguments because the notion of perfection plays a significant role. It is similar to both Gӧdel and Leibniz’s arguments because it attempts to demonstrate that the concept of God (in this case, a supreme being) is possible in addition to demonstrating God’s existence.

Maydole’s argument is not that much different from other modal ontological arguments, though the formal logical statement of his argument is much more rigorous than that given by many others. There is a move from it being possible that a supreme being exists to a supreme being existing, and the rest of the argument is in service to showing that a supreme being is possible.

There is one other difference worth mentioning. Maydole’s argument does not set out to prove that God necessarily exists, just that God exists. This is different than most modal ontological arguments since most modal ontological arguments aim for

God’s necessary existence – indeed, that is one of the defining features of the class of modal ontological arguments.

Section X. Leibniz’s Argument

The most straightforward version of the ontological argument given by Leibniz is in “Two Notations for Discussion with Spinoza,” (Leibniz 1970). As with all of our

20 other authors, there are a number of interpretations of exactly how to formulate and analyze the arguments given, and many look to a number of other works in Leibniz to get a full grasp of the ontological argument. For the purposes of this chapter, we will examine primarily the argument found in the Two Notations. The argument proceeds as follows.

1. A perfection is a simple quality that is positive and absolute. 2. God is defined as a being possessing all perfections. 3. Existence is a perfection. 4. All perfections are compossible (they can all be possessed at the same time). 5. God is possible. 6. If the concept of God is possible, then it exists (since the concept contains the perfection of existence). 7. Therefore, God exists.

If a perfection is simple, then it cannot be demonstrated to be incompatible with another perfection, since any demonstration will require analysis of the perfection and a simple is, by definition, unanalyzable. If that is true, then all perfections are compossible. That is merely another way of saying that God is possible. Since existence is a perfection, then as long as the concept of God is possible, then God will exist. So God exists.

Leibniz also gives a version of the argument focusing on necessary existence instead of simply existence. After demonstrating that all perfections are compossible and that existence is a perfection, Leibniz attempts to ensure that the kind of existence is necessary existence. He does this by attempting to show that what we mean for a being to necessarily exist is just for existence to belong to the essence of a being. “Again, a necessary being is the same as a being from whose essence existence follows. For a necessary being is one which necessarily exists, such that for it not to exist would imply

21 a contradiction, and so would conflict with the concept or essence of this being.”

(Leibniz 1992).

Section XI. Conclusion

In this chapter, I have forecast each later chapter of the dissertation and explained the structure of the dissertation as a whole. I have examined a number of important versions of the ontological argument, both historical and contemporary. Each of these arguments is committed to what I will call a Linkage Premise, which is a premise that there is some tight connection between goodness, or greatness, or perfection, and existence (either existence in general or necessary existence). My central argument is that any ontological argument that requires a Linkage Premise is committed, either explicitly or implicitly, to the convertibility thesis of being and goodness. In order to evaluate this claim, we will need to examine the convertibility thesis of being and goodness in some detail, which occurs in chapter 4. However, we will also need to evaluate whether or not these arguments require any particular account of existence in their Linkage Premise, as Kant and others have argued. This issue is taken up in the next chapter.

22

Chapter 2: Senses of Existence

Section I. Introduction

A major concern for any philosopher engaging with the ontological argument for the first time is the nature of existence. How do we understand phrases such as

“existence is a perfection,” “existence is a property” or their denials? What does it mean to say that God necessarily exists? How does this relate to other fundamental distinctions in philosophy, such as that between accidental and essential properties, or between substances and attributes? This chapter will not attempt to answer all of those questions. In this chapter, I will begin by discussing the Kantian line of critique that the ontological argument commits a mistake by supposing that existence is a predicate or property. I will examine a very common, contemporary view of existence, most appropriately traced back to the works of Gottlob Frege and Bertrand Russell in support of this critique. Then, I will consider alternative accounts of existence that allow for existence, in some way, to apply to individuals and objects, as found in the works of

Terence Parsons and Peter Geach. After this, I will discuss Norman Malcom’s contribution that “existence” and “necessary existence” differ in their ontological import, and thus modal versions of the ontological argument need not worry about this traditionally Kantian line of resistance. This chapter is important as it ends with two important considerations. First, that on some of these alternative views of existence, it is not incorrect to say that existence can be a property of an object – so the critique that existence is not a property or at least not a property of objects full stop (the Kantian critique) is no longer a viable objection to the ontological argument. Second, that even if existence itself is not a predicate, this does not harm the ontological argument,

23 whether because of some significant difference between necessary and contingent existence, or because of alternative modal formulations of the argument that do not require that existence be a property of objects. The upshot is that in any case, ontological arguments can proceed without requiring one to adopt a specific position on the intricate, abstract, and opaque debate regarding the nature of existence.

Section II. Existence Is Not a (First-Order) Predicate or Property

In order to see how the Kantian critique is supposed to apply to the ontological argument, it is important to explicate what is meant by the sentence “existence is not a predicate.” Although this is the most popular way to word the critique, it is imprecise.

The critique is best understood as “existence is not a first-order/first-level/real predicate or property.” What this means takes on slightly different characters in different writers, but all agree with something like the following. The notion of existence is captured by the existential quantifier in propositional logic. Because of this (or by giving a certain gloss of this), existence is not a property of individual objects and cannot be applied, as a predicate, to individual objects. To think of or consider a concept is to already consider it as if it were existing – we determine which things exist by restricting or expanding the scope of our domain of discourse.

To understand the import of this conception of existence in the contemporary philosophical world, I examine Frege and Russell‘s contributions to this discussion.

Since Frege’s view of existence is carried on in a slightly different way by Russell, the views on existence I will discuss in this section are broadly Fregean, although there are some slight differences in the two views. I also point out that since the heavy lifting is

24 done by Frege’s initial formulation of this thesis, I will spend less time explicating

Russell’s views.

II.A Frege

Frege’s account of existence can be best understood by examining Frege’s understanding of concepts as related to objects and to other concepts. In an atomic sentence, such as “God is omnipotent” we have a subject, “God,” and the first-order concept “omnipotent,” which tells us about or applies to the subject, the individual object “God.” So far, this view has similarities with other traditional accounts of the relationships between subjects and predicates or entities and their attributes. For Frege,

“omnipotence” does not itself entail some object or existent entity within the world, but describes some other thing, the subject to which it applies. So far, then, we have individual objects, and first-order concepts that can be rightfully said to apply to or describe or tell us about those objects.1

In addition to first-order concepts, we also have second-order concepts. Second- order concepts apply to first-order concepts. Both the concept of existence and numerical concepts are second-order concepts according to Frege. For example, when thinking of the number one, Frege has the following to say;

If, for example, we collect under a single concept all concepts under which there falls only one object, then oneness is a component characteristic of this new concept. Under it would fall, for example, the concept “moon of the Earth,” though not the actual heavenly body called by this name. In this way, we can make one concept fall under another higher or, so to say, second-order concept. This relationship, however, should not be confused with the subordination of species to genus. (Frege 1960, p.65)

1 There is a wrinkle here. Singular expressions name or refer to objects, but are not themselves objects. So “God” is a singular term, a naming term, that refers to or a picks out the individual object God. Since this additional detail does not affect my representation of Frege’s view (or at least the point I am expressing) I do not go into further detail here. 25

This can also be brought out by considering another of Frege’s remarks. He writes

Further confirmation of the view that number is assigned to concepts is to be found in idiom; just as in English we can speak of ‘three barrel,’ so in German we speak generally of ‘ten man,’ ‘four mark’ and so on. The use of the singular here may indicate that the concept is intended, not the thing. The advantage of this way of speaking is particularly noticeable in the case of the number 0. Elsewhere, it must be admitted, our ordinary language does assign number not to concepts but to objects; we speak of ‘the number of bales’ just as we do of ‘the weight of the bales.’ Thus on the face of it we are talking about objects, whereas really we are intending to assert something of a concept. This usage is confusing. The construction in ‘four thoroughbred horses’ fosters the illusion that ‘four’ modifies the concept ‘thoroughbred horse’ in just the same way as ‘thoroughbred’ modifies the concept ‘horse.’ Whereas in fact only ‘thoroughbred’ is a characteristic used in this way; the word ‘four’ is used to assert something of a concept. (Frege 1960, p. 64, emphasis mine)

Although we can consider “thoroughbred” a first-level predicate that applies to the object “horse,” we cannot understand the numerical concept “four” to apply in the same way. We may be tempted to think that numbers apply to individual objects, as do first- order concepts, but in understanding the relationship between numbers and other concepts, we find that numbers actually apply to concepts and not to objects, and therefore that numbers are second-order concepts.

What do we mean when we talk about existence? For Frege, interestingly enough, this question can be answered not only by considering the role of existence in relation to atomic and complex sentences, but also by considering the relationships between existence and the number one, and nonexistence and zero, or the number nought (this is Frege’s terminology). Just as the concept of the number one is a second- order concept under which we may organize or classify first-order concepts (or in

26

Frege’s language, under which first-order concepts fall), so too with existence and its relationship to other first-order concepts. Frege writes

…the proposition that there exists no rectangular equilateral rectilinear triangle does state a property of the concept ‘rectangular equilateral rectilinear triangle’; it assigns to it the number nought. In this respect existence is analogous to number. Affirmation of existence is in fact nothing but the denial of the number nought. Because existence is a property of concepts the ontological argument for the existence of God breaks down. (Frege 1960, p. 65)

Since existence is a property of concepts, a second-order concept, then it cannot be applied to individual objects, in the same way that numerical concepts cannot apply to individual objects.

In summary, Frege understands existence to be a second-order concept, and thus is plainly inapplicable when predicated of an individual object, whether that individual object is Rob Byer, the computer in front of me, or God.2 The concept existence says of the first-order concept it is applied to that there is at least one object that is a member of that concept set or class; another way to put it (perhaps unhelpfully) is that at there is at least one instantiation of the first-order concept. Again, then, existence applies to first- order concepts, and not individual objects. It is nonsense on Frege’s view to say that an individual object does or does not exist.

Hence, what is here asserted about a concept can never be asserted about an object... I do not want to say it is false to assert about an object what is asserted here about a concept; I want to say it is impossible, senseless, to do so. The sentence ‘there is Julius Caesar’ is neither true nor false but senseless… (Frege 1980a, p. 175)

II.B Russell

2 Frege explicitly explains that he takes this distinction to illuminate a fallacy committed by the ontological argument in Frege 1960 and Frege 1980b. 27

Initially, Russell does seem to countenance two senses of existence; one that is used in philosophical questions and in everyday life, and the other that is used by logicians in the form of the existential quantifier (Russell 1905, 398). During later formulations of his views, though, Russell uses the logician’s sense of existence to the exclusion of the everyday conception.

Russell’s view is really a development of Frege’s view, and as such shares the same basic positions on the question of existence. He phrases things differently; existence is a matter of whether or not a propositional function can be said to be true.

“When you take any propositional function and assert of it that it is possible, that it is sometimes true, that gives you the meaning of existence. You may express it by saying that there is at least one value of x for which that propositional function is true,”

(Russell 2010, 195). For Russell, then, existence is to be understood by fleshing out some account of modal and logical terms. If what it means for something to exist is that for the propositional function being considered, there is some assignment of values that makes the statement true. For example, he writes, “One may call a propositional function necessary, when it is always true; possible when it is sometimes true; impossible when it is never true,” (Russell 2010, 193 and 195). We’ve already seen from the earlier statement that the matter of existence is about the truth value of a propositional function when considering all, or some variables. He states it most clearly by saying “Existence is essentially a property of a propositional function. It means that the propositional function is true in at least one instance,” (Russell 2010, 195).

Now that we have sketched Russell’s account of existence, we can ask how, if at all, he differs from Frege in determining what kinds of entities the concept “existence”

28 applies to. In short, we find that he adopts the exact same position as Frege; existence can be said of a class of things, or of a set of things, but never of individual things. He expresses this similarity best in the following passage, which I quote at length.

Therefore when you say ‘Unicorns exist,’ you are not saying anything about any individual things and the same applies when you say ‘Men exist.’ If you say that ‘Men exist, and Socrates is a man, therefore Socrates exists,’ this is exactly the same sort of fallacy as it would be if you said ‘Men are numerous, Socrates is a man, therefore Socrates is numerous’ because existence is a predicate of a propositional function, or derivatively of a class. When you say of a propositional function that it is numerous, you will mean that there are several values of x that will satisfy it… If x, y, and z all satisfy a propositional function, you may say that that proposition is numerous, but x, y, and z severally are not numerous. Exactly the same applies to existence, that is to say that the actual things that there are in the world do not exist, or, at least, that is putting it too strongly because that is utter nonsense. To say that they do not exist is strictly nonsense, but to say that they do exist is also strictly nonsense. It is of propositional functions that you can assert or deny existence… If I say ‘The things that there are in the world exist’ that is a perfectly correct statement, because I am there saying something about a certain class of things; I say it in the same sense in which I say ‘Men exist.’ But I must not go on to ‘This is a thing in the world, and therefore this exists.’ It is there the fallacy comes in, and it is simply, as you see, a fallacy of transferring to the individual that satisfies a propositional function a predicate which only applies to a propositional function. (Russell 2010, 196-197)

Russell’s main point in the passage quoted above is exactly the same as Frege’s point in both Frege 1960 and Frege 1980a regarding applying the concept of existence to individuals; it is simply an incorrect usage of the predicate “existence.”

For Russell then, as for Frege, the slogan “existence is not a predicate” amounts to the idea that existence is only applicable to classes of things, or propositional functions. Applying the concept to individual objects or entities is to misunderstand the notion of existence. It is just as much nonsense on Russell’s view to say “Alvin

29

Plantinga exists” as it is to ponder the question of whether or not God exists. Existence is simply not a concept or a property that can apply to individual things.

II.C Commonalities

When I refer to common, contemporary views of existence related to Frege and

Russell, I do not mean that every contemporary metaphysician accepts Frege’s view about concepts and objects exactly, or Russell’s view about definite descriptions and paraphrases, I only mean that, whatever else is required when discussing existence and what it means, I believe that many contemporary accounts of existence follow this form.

In particular and with respect to the central concern of the dissertation (the ontological argument), I believe that this exact style of reasoning is used when philosophers give arguments against using existence as a predicate or property of God in the ontological argument.

On both views, a) existence is something that can only be predicated of either first-order concepts or propositional functions or classes of things, and cannot apply to individual objects or entities, and b) existence is univocal, including any supposed distinction between “being” and “existence.” If we understand by “God” an individual object, with its own identity conditions and thus distinct from other objects or from the world as a whole, then it is a mistake to talk of predicating “existence” of God in just the same way as it would be a mistake to talk of predicating “numerous” or “one” of

God. In other words, the ontological argument rests upon a category mistake. Of course, on the views given above, any ordinary talk about an object’s existing will be equally meaningless and equally rest on a category mistake, which means that this is not a problem unique to God. This kind of view would really rule out all ordinary

30 assertions about the existence of objects (or at least require them to be suitably rephrased).

The other commonality that this view shares with other common contemporary views of existence is that existence is univocal, that there is no distinction between

“being” and “existence” or different senses of the word “exists.”3 Even in cases of other disagreements, such as whether or not existence can apply to objects (van

Inwagen for example thinks that it can), these views all deny a distinction between either “being” and “existence” or different senses of the word “exists.” This is important because the alternative accounts offered below do treat existence as if it can have different meanings, and it is an open question whether or not we should treat existence and being as if they are simply synonyms from some other carry-over of language or if they express different, if related, notions. This would give us a second direct conflict between the accounts listed above and those listed below.

Section III. Existence Is a (First-Order) Predicate/Property

The strategy in this section of the chapter is to outline two different theories of existence that either allow for existence to be used as a first-order predicate and to be applied to individuals, or allow existence to be used differently than in the Fregean tradition. Outlining these systems responds to the two major commonalities discussed above; that existence cannot be applied to individual objects or entities, including God, and that the concept of existence is univocal. Proponents of the systems given above

(including contemporary philosophers) have often taken it to be the case that the ontological argument fails because it utilizes an incorrect notion of existence; philosophers of religion and theologians who promote the ontological argument are just

3 This is true of Quine 1948 and van Inwagen 2014. 31 confused about how the term “existence” works. If either of the systems given below is an accurate account of existence, then critics of the ontological arguments need to rethink the Kantian line of critique. I will first examine Parson’s account of existence, and then Geach’s. Finally, at the end of this section I examine the remarks of Norman

Malcolm in “Anselm’s Ontological Arguments” about necessary existence as opposed to mere existence. The upshot here is twofold: either a system that allows “existence” to be used as a first-order predicate, or to apply to individual objects, or a suitable distinction between the applicability of the concepts of “existence” and “necessary existence” can allow a successful ontological argument.

III.A Parsons

In Nonexistent Objects (Parsons 1980), Terence Parsons argues for and develops a Meinongian theory of objects that allows us to talk about and discuss nonexistent objects and fictional objects without falling into the trap of the negative existential proposition, and that preserves some straightforward ways of talking about these objects. Parson distinguishes between what is meant by existential quantification and what is meant by saying that something exists. Strictly speaking, then, Parson’s does not distinguish two different uses of the term “exists” as does Geach, but rather differentiates statements of the form “There is a table” (the quantification statement) and statements of the form “A table exists,” (the existence as a property statement).

Parsons framework allows us to separate the existential quantifier from the notion of existence, and elements of his view support the idea that existence, in some important sense, can apply to individual objects.

32

As with Frege, and Russell, it is beyond the scope of this paper to do a full and thorough analysis of the whole of Parsons’ system. Luckily we need not dig into

Parsons detailed treatments of fictional and mythical objects, or the application of his theory to monads in order to bring out the features of the theory that repudiate the

Fregean views given above. For our purposes, it will be enough to discuss the distinction between nuclear and extranuclear predicates and corresponding nuclear and extranuclear properties, the distinction between “being existent” and “existing” (or between statements of the form “There is a table” and statements of the form “A table exists”), and Parsons’ remarks on the applicability of his system to the ontological argument.

First, we must distinguish nuclear and extranuclear predicates, and the easiest way to proceed is to give examples and then analyze them. Nuclear predicates are those predicates we normally think of as specifying a property that applies to individual objects in an appropriate way. For example Parsons lists the following as examples of nuclear predicates: “is blue,” “is tall,” “kicked Socrates,” “was kicked by Socrates,”

“kicked somebody,” “is golden,” and “is a mountain,” (Parsons 1980, 23). These are all predicates that represent a set of properties had by some objects. Extranuclear predicates, on the other hand, do not apply to individual objects. Extranuclear predicates fulfill one or another function in Parsons’ system and in ordinary discourse, depending upon the extranuclear predicate.4 Examples of extranuclear predicates

4 Concerning whether or not there are such things as extranuclear properties, Parsons writes “I think it is not very important whether you say that extranuclear predicates stand for a special sort of property— extranuclear properties—or whether you say that they do not stand for any properties at all. In the theory I sketch, I will assume that there are extranuclear properties; this allows me a freedom of exposition I would not otherwise have.” Parsons 1980, 26. I will assume, along with Parsons, that there are such things as extranuclear properties. 33 include the following categories and corresponding examples: the ontological predicates such as “exists,” “is mythical,” “is fictional,” the modal predicates such as “is possible,”

“is impossible,” the intentional predicates such as “is thought about by Meinong,” “is worshipped by someone,” and the technical predicates such as “is complete,” “is incomplete,” “is logically closed,” (Parsons 1980, 23).5 Parsons is less than clear about what exactly nuclear predicates and extranuclear predicates are, but he does discuss a decision procedure by which we distinguish nuclear and extranuclear predicates and how this distinction relates to the Fregean views discussed above. Concerning the distinction between nuclear and extranuclear predicates, he writes

Our historical situation yields a very rough kind of decision procedure for telling whether a predicate is nuclear or extranuclear. It is this: if everyone agrees that the predicate stands for an ordinary property of individuals, then it is a nuclear predicate and stands for a nuclear property. On the other hand, if everyone agrees that it doesn’t stand for an ordinary property of individuals (for whatever reason), or if there is a history of controversy about whether it stands for a property of individuals, then it is an extranuclear predicate and it does not stand for a nuclear property. (Parsons 1980, 24)

Concerning the relationship of the distinction between nuclear and extranuclear predicates and Fregean views, he writes

I’d like to emphasize that this division of predicates into nuclear and extranuclear is not peculiar to Meinong at all; it’s an old and familiar one. People such as Frege and Russell distinguish predicates that stand for properties of individuals from those that don’t. The extranuclear predicates listed above are mostly ones that Frege and Russell have been telling us all along do not stand for properties of individuals. (Parsons 1980, 23-24)

So far, this view does not immediately lend itself to a different view of existence than those espoused by Frege and Russell, especially as Parsons classifies “existence”

5 The last two examples of technical predicates are my addition, but allow us to flesh out the category of technical predicates further. 34 as an extranuclear predicate and states that extranuclear predicates do not stand for properties that can be applied to individual objects. In order to see if this system allows for even the possibility of an ontological argument, we must examine Parsons’ discussion of “being existent” as opposed to “existing” or put another way, the difference between “is” and “exists.”

In considering Russell’s criticisms of Meinong, Parsons considers whether there is any difference between “being existent” and “existing.” For example, if we have the concept of “the existent golden mountain,” does this mean the golden mountain exists?

According to Parsons, the answer to this question is a simple “No.” The golden mountain “is existent” but does not “exist.” If this is the case, though, it seems that the predicate “being existent” can apply to individual objects, and if this is correct, it is a nuclear predicate and stands for a nuclear property, unlike “exists,” (Parsons 1980, 42).

The important question that arises at this point is “what is the relationship between

‘being existent’ and ‘exists’?” Parsons argues that at least some extranuclear properties have watered-down nuclear versions (Parsons 1980, 44). “Being existent” is the watered down nuclear version of the extranuclear property “exists,” (Parsons 1980, 43-

44). If the foregoing is correct, then “being existent” can apply to individuals, and thus to God. Is this helpful? It depends on what it is we are talking about when we talk about “being existent.” Parsons writes

Given that there are nuclear properties true of all existing objects, are any of them denoted by ‘being existent’?... since the controversy over whether existence is a property uses the word ‘existence,’ and since both sides of the dispute seem to have some claim to plausibility, it seems reasonable to hold that the reason the dispute goes on is that both sides are in some sense right. And this would be explained if ‘exists’ or its cognates were ambiguous, standing for both a nuclear and an

35

extranuclear property, neither of which alone exhausts the way in which we use the word. (Parsons 1980, 43-44)

If this is correct, then there is a sense of “existence” that stands for a nuclear property that can be said to hold of objects, in addition to the extranuclear or second-order sense of “existence” as discussed by Frege and Russell.

But what does this mean for the ontological argument? Parsons allows that it is possible for God to exist in his system. He writes “The present system is so rich in objects that one might suspect that God would be found among them. That depends on what is meant by God… If ‘God’ means, say, ‘the existing deity of the Bible,’ then there may or may not be such an object, and the quoted definite description may or may not refer…” (Parsons 1980, 212). He goes on to say that the argument will probably

“rely on fallacious reasoning… The only support it receives from the theory of objects is the vindication of the coherence of reasoning about a thing without prejudice as to whether or not that thing exists,” (Parsons 1980, 215). Though Parsons does not find the argument convincing, I think there is another way in which the theory of objects outlined by Parsons supports the possibility of an ontological argument. In considering reasons to reject the argument, he discusses the Fregean criticism as follows.

According to Frege, existence isn’t a property of individuals, but rather a property of concepts; existence is not properly represented in a language by a predicate but rather by a quantifier. However, there certainly is a property of individuals represented by the complex predicate ‘being something such that there exists such a thing’, and the arguments can all be rephrased (though perhaps clumsily) in terms of this predicate. (Parsons 1980, 216, emphasis mine)

Since there is a version of the predicate “existence” that is a nuclear predicate, and represents a nuclear property that can be applied to individuals, then it seems we can apply such a predicate to God. It also seems that Parsons recognizes, in the bold section

36 of the above quote, that in his system there is a way to re-phrase the argument such that it is an open question of whether or not God exists (clumsy though he thinks it might be). If the foregoing is true, then the Kantian inference is blocked.

III.B Geach

I start this section by noting that although Geach does think that there are two senses of existence, one corresponding to the orthodox Fregean view given above, and another that can properly be said to be predicated of individuals, at times he does not think the latter sense is what is under consideration when we say “God exists,” (as in

Geach 1955). However, in other cases there is a sense of existence, suitably predicated of individuals, that does apply to God (as in Geach 1968). I will try to explain these different views by explaining how Geach understands the different senses of existence in both “Form and Existence” (Geach 1955)6 and in “What Actually Exists” (Geach

1968). I will end by arguing that if Geach is correct, the Kantian inference is blocked.

In Geach 1955, Geach denies that individual usage of the term (in C propositions as discussed in the next paragraph) can apply to God, and it is beyond the scope of this section to argue with Geach on this matter. For this section, I am only concerned to show that against Frege and Russell, existence can be said to apply to individuals. At least at the most basic, conceptual level, it seems straightforward to say that if existence can be properly said of individuals, and if God is an individual, then it is possible to predicate existence of God, and contra Geach, the ontological argument is not a

“fallacy,” (Geach 1955, 268)

6 In this paper, Geach is giving an explanation of a view developed by Aquinas, but based on his discussion in it and in Geach 1968, I believe it is fair to treat this as Geach’s view as well. 37

In discussing different kinds of existential propositions, Geach uses negative existential propositions to bring out the difference in the kinds of existential propositions (Geach 1955, 262 – 264). Geach says that of existential propositions, there are three kinds: A propositions, such as “Cerberus does not exist,” B propositions such as “Dragons do not exist,” and C propositions such as “Joseph is not (or Joseph does not exist),” (Geach 1955, 264). Of the first kind of existential proposition, Geach says that

“in A propositions, ‘exists’ or ‘is real’ is not a predicate, not even of the word

‘Cerberus.” In the subsequent discussion, Geach states that in A propositions, existence seems to be predicated of a named subject. Uses of “existence” and the denial of it in A propositions always apply to singular, named individuals. When we deny that something, like Cerberus exists in this sense, we are saying that the word fails to refer at all. As Geach puts it,

“To show the real force of the parent's reassuring "Cerberus does not exist", and how it is about the word "Cerberus", we cannot content ourselves with writing "(The word) 'Cerberus' does not exist", but must completely recast the sentence, say as follows: "When I said ' Cerberus ' in that story, I was only pretending to use it as a name" (Geach 1955, p. 264)

In B propositions, existence is meant only to say that a thing is predicated (of a class of objects). In relation to B propositions, he writes “’an F exists’ is true if and only if ‘F’ is truly predicable of something or other… in predicating ‘God exists’ we are not predicating something of God, but predicating the term ‘God’ itself; ‘God exists’ means ‘something or other is God.’” (Geach 1955, p. 266). It looks as if on Geach’s account, in both A and B existential propositions, existence is not a predicate that applies to individual objects, but something closer to a second-order concept in the

Fregean sense. It is important to note that Geach also says that when we speak of the

38 claim “God exists” we are using the B proposition, and not either of the two others

(Geach 1955, p. 266). This is why he believes the ontological argument to rest on a fallacy. Since B propositions do not allow existence to be predicated of individuals, and if B propositions are how we are to understand the claim that “God exists” then it appears that we do not predicate existence of God when we claim that “God exists,” we are instead asserting something like “The unit-set God is not empty.”7 And if this is all we do when we argue that God exists in the ontological argument, we are in double- trouble – the concept we seem to be applying to an object cannot do that job, and the true meaning of the phrase is to simply beg the question that the argument is purporting to answer.

In considering C propositions, Geach seems to understand existence as predicable of individuals. “We have here a sense of ‘is’ or ‘exists’ that seems to me to be certainly a genuine predicate of individuals; the sense of ‘exist’ in which one says that an individual came to exist, still exists, no longer exists, etc.,” (Geach 1955, 266 –

267) On this view, C propositions do express ‘existence’ as being predicable of individual objects. In this kind of proposition, “existence” always involves duration.

The idea here is that in some cases of asserting that an entity exists we are locating it temporally. The examples he gives are, as stated above, “Joseph is not and Simeon is not,” (Geach 1955, 266). In this case, we are talking about an object’s either possessing some property (existence in C propositions) or lacking it. However, this seems to relate to existence at a time or in time. In discussing whether or not existence in C propositions could be uttered or denied of God, Geach suggests that

7 I thank Neal Judisch for this way of phrasing Geach’s statement. 39

Some people may not easily see the difference between "God exists” (sense B) and “God is” (sense C). But in the contradictories the difference is apparent. "God is not" (sense C) would have to be construed like "Joseph is not " ; it would then suitably express the supposition that perhaps the world was made by an old superannuated God who has since died (a suggestion of Hume's). This is quite different from the atheist's “there is no God” (sense B). One would indeed wish to say that everlasting existence is part of the concept of a God; of Hume's senile creator one would wish to say that since he is dead now he never was God when he was alive. But saying this does not commit us to the fallacy of the Ontological Argument. It belongs to the concept of a phoenix that it should never die by accident and should ward off old age with a bath of flames at regular intervals; so a dead phoenix is a contradiction in terms. But this does not mean that there must be a live phoenix. If there is a God, then he lives for ever; but we cannot determine from this whether there is a God. (Geach 1955, p. 268)

What, exactly, is the sense of existence here? There are at least two things going on: 1) that C-existence is about time and continued or continuing existence, and

2) that this sense of existence comes very close to capturing the idea of being alive, and its denial of being dead. Although this does not help the ontological argument by itself, as Geach suggests, it does serve to undermine the Kantian objection. It looks like in this sense C, Geach defends a two-sense view of existence; we can predicate

“existence” of individuals!8

When we talk about God’s existence, according to Geach 1955, we are using the

B sense of existence. Recall, the B sense of existence is whether or not some predicate can be appropriately applied at all. To say, along with Geach, that it is true that “An F exists” is just to say “Something or other F’s,” (Geach 1955, p. 266). If this is what we mean when we talk about God, it will be inapplicable as described above. On the other hand, in “What Actually Exists” (Geach 1968) the matter becomes a bit more complicated. Geach there distinguishes between the classical Fregean notion of

8 In this section, Geach explains that “God” is not supposed to be a logically proper name, but instead a “descriptive, predicable, term,” in Geach 1955, p. 266. 40 existence as a second-order concept, and existence as actuality. He writes “Existence in the sense of actuality is several times over emphatically distinguished in Frege’s works from the existence expressed by ‘there is a so-and-so’… Actuality is attributable to individual objects; the existence expressed by ‘there is a—‘ is not,” (Geach 1968, 7).

To clarify what is meant by actuality, Geach writes “x is actual if and only if x either acts, or undergoes change, or both; and here I count as ‘acting’ both the inner activities of mind, like thinking and planning, and the initiation of changes in things,” (Geach

1968, 7). If we think of this sense of existence in relation to the existence expressed by

C propositions, we may have two forms of existence that can be applied to individuals: existence as a tensed predicate of their coming to be, or not existing, etc., and existence as actuality, where to predicate actual existence or actuality of a thing it is sufficient to hold that “x either acts, or undergoes change, or both.” I do not know if Geach holds that these two senses of existence are distinct, but nothing he says indicates that he thinks of them this way. However, Geach thinks that we can rightfully apply existence as actuality to God (Geach 1968, 15 – 16). If this is true, then there is a sense of existence as a predicate that applies to individual objects, which can indeed apply to

God.

So, is there a sense of existence that applies to individual objects and that would thereby apply to God? As noted above, in Geach 1968 the answer is yes. For the purposes of this section, I have primarily concerned myself with showing that for

Geach, existence can be properly predicated of individuals. This is important since the major driving force behind the Kantian critique, and supported by the more recent views of existence of Frege and Russell, is that the ontological argument rests on a category

41 mistake about whether it is even possible for existence to be a property of an individual object. Insofar as Geach’s answer is positive, this will remove the viability of this overarching criticism.

Although I pushed back against this, our immediate prior section on Parsons ended with Parsons suggesting the ontological argument would get no or little help from his system; there, as with my remarks on Geach in this section, I am primarily focused on whether or not the resources of the system block the Kantian critique, and I believe that both Parsons’ and Geach’s systems offer the resources to do just that.

III.C Malcolm

In defending the modal version of Anselm’s argument, Malcolm examines the familiar Fregean line of critique and some of its progeny. Malcolm attempts to buttress the viability of the argument against these critiques by discussing the difference between “existence” and “necessary existence,” and how we think these terms apply.

Whether or not we can predicate “existence” of God, Malcolm argues that that we can certainly predicate “necessary existence” of God.

What does it mean to say that “God necessarily exists?” For Malcolm, it can be brought out by considering God’s unlimited nature. Malcolm links our understanding of inferiority, dependence, and contingency and similarly examines the relationships between superiority, independence, and necessity (Malcolm 1960, 46 – 48).

God is usually conceived of as an unlimited being. He is conceived of as a being who could not be limited, that is, as an absolutely unlimited being. This is no less than to conceive of Him as something a greater than which cannot be conceived. If God is conceived to be an absolutely unlimited being He must be conceived to be unlimited in regard to His existence as well as His operation. In this conception it will not make sense to say that He depends on anything for coming into or continuing in existence… (Malcolm 1960, 47)

42

I think this is the correct way to approach the question of whether or not we can predicate existence of God. Malcolm rightly notes that some philosophers think it is nonsense to talk about necessary existence at all. We may freely speak of necessary properties of other kinds, such as the necessary property of “being an unmarried male” when discussing the term “bachelor” or that “a triangle necessarily has three sides.”

But existence is not the kind of property (if it is a property at all—recall Frege and

Russell’s remarks) that can be said to have the modal status of “necessary.” Can we predicate necessary existence of God? Malcolm answers in the affirmative.9

…The view that logical necessity merely reflects the use of words cannot possibly have the implication that every existential proposition must be contingent. That view requires us to look at the use of words and not manufacture a priori theses about it. In the Ninetieth Psalm it is said: ‘Before the mountains were brought forth, or ever thou hadst formed the earth and the world, even from everlasting to everlasting, thou art God.’ Here is expressed the idea of the necessary existence and eternity of God, an idea that is essential to the Jewish and Christian religions. In those complex systems of thought, those "languages-games," God has the status of a necessary being. Who can doubt that? Here we must say with Wittgenstein ‘The language game is played!’ (Malcolm 1960, 53-55)

I think it is established that a proponent of the argument can predicate necessary existence of God; any proponent of contemporary modal ontological arguments will affirm this. I think that this cuts to the heart of the matter most fantastically. Whether or not we can apply the concept of existence to God, there seems to be no trouble applying the concept of necessary existence to Him. Even critics of the argument say that proponents can do so, but they disagree with those same proponents about the

9 See also Plantinga 1974, among others who have held that it makes sense to talk about God as necessarily existent. 43 import of such predication.10 With regards to whether or not when we say “God necessarily exists” we mean something like “If God exists, then He necessarily exists,”

I am inclined to agree with Malcolm.

I think that Caterus, Kant, and numerous other philosophers have been mistaken in supposing that the proposition ‘God is a necessary being’ (or ‘God necessarily exists’) is equivalent to the conditional proposition ‘If God exists then He necessarily exists.’ For how do they want the antecedent clause, ‘If God exists,’ to be understood? Clearly they want it to imply that it is possible that God does not exist. The whole point of Kant's analysis is to try to show that it is possible to ‘reject the subject.’ Let us make this implication explicit in the conditional proposition, so that it reads: ‘If God exists (and it is possible that He does not) then He necessarily exists.’ But now it is apparent, I think, that these philosophers have arrived at a self-contradictory position. I do not mean that this conditional proposition, taken alone, is self-contradictory. Their position is self-contradictory in the following way. On the one hand, they agree that the proposition ‘God necessarily exists’ is an a priori truth; Kant implies that it is ‘absolutely necessary,’ and Caterus says that God's existence is implied by His very name. On the other hand, they think that it is correct to analyze this proposition in such a way that it will entail the proposition ‘It is possible that God does not exist.’ But so far from its being the case that the proposition ‘God necessarily exists’ entails the proposition ‘It is possible that God does not exist,’ it is rather the case that they are incompatible with one another! Can anything be clearer than that the conjunction ‘God necessarily exists but it is possible that He does not exist’ is self-contradictory? Is it not just as plainly self-contradictory as the conjunction ‘A square necessarily has four sides but it is possible for a square not to have four sides?’ In short, this familiar criticism of the ontological argument is self-contradictory, because it accepts both of two incompatible propositions. (Malcolm 1960, 57 – 58)

This seems to me to be the most powerful part of Malcolm’s defense of Anselm. I think many contemporary philosophers understand the phrase “God necessarily exists” as being translatable to “If God exists, He necessarily exists.” Further, at least in the

Anselmian tradition we must speak of God as a necessary being, and even figures such as Oppy and Findlay are comfortable speaking this way. They claim that for different

10 For example, see Oppy’s remarks regarding the proper discharging of operators in modal ontological arguments involving necessity in Oppy 1995, 72-74, and his “General Objection” in 114-118. 44 reasons, the argument fails to establish the existence of God, much less the necessary existence of God, but they are comfortable in talking about the conception of God as a necessary being. In a certain sense they should be so comfortable; Hartshorne and

Malcolm both affirm that God’s modal status is necessary, and thus that either God is necessarily existent, or necessarily nonexistent, and thus impossible. One need not be a fanatical theist to understand the concept of God in this way, but when opponents of the argument admit this much, it brings out the exact self-contradiction that Malcolm is talking about in the lengthy passage given above. Ultimately, I think that Malcolm is right when he says “once one has grasped Anselm's proof of the necessary existence of a being a greater than which cannot be conceived, no question remains as to whether it exists or not, just as Euclid's demonstration of the existence of an infinity of prime numbers leaves no question on that issue,” (Malcolm 1960, 52). I do not mean to say that no one can consider Anselm’s argument, or better, a modal ontological argument without becoming convinced that there necessarily exists a God. But I do think that it leads us to the proposition expressed by Malcolm and Hartshorne, that God’s existence is either necessary or impossible. Failing a good reason to think that the concept is self- contradictory or metaphysically impossible for some other reason, and if we indeed think that the concept is possible, then there is no longer any question on the matter.11

Section IV. Concluding Dilemma and Concluding Argument

11 This is, to put it mildly, a contentious and difficult issue. There has been a range of interesting work done on the relationship between conceivability and possibility, but it would push us too far afield to consider this issue at this point. I am interested in pursuing this matter in later research, as I believe that ultimately an important consideration of many philosophical arguments, including the ontological argument, is the gap between conceivability and possibility, and different models of the relationship between these concepts. Interested readers should refer to the essays in Gendler and Hawthorne 2002a, especially Chalmers 2002, Gendler and Hawthorne 2002b, Yablo 2002, as well as articles outside of that collection including Chalmers 1999, Chalmers 2006, Vaidya 2015, van Inwagen 1998, Yablo 1993, and Yablo 1999. This is far from a complete or representative list of the vast and exploding literature on this topic, but I have found these articles to be helpful in developing my own understanding of the topic. 45

I conclude this chapter by giving a dilemma.

1. Either existence is a property of objects, or it is not a property of objects (whatever else it may be a property of). 2. If existence is a property of objects, then the Kantian objection is false. 3. If the Kantian objection is false, then ontological arguments will not fail for that reason alone. 4. If existence is not a property of objects, ontological arguments may still be made without engaging in any fallacious reasoning specific to whether or not existence is a property of objects. 5. So, whether existence is a property of objects or not, the ontological argument will not fail on the basis of considerations of the nature of existence by itself. Premise 1 is simply an exhaustive and exclusive disjunction. Either existence is a property of objects, or it is not. No significant defense needs to be made in service of this premise.

Kant and others have expressed the idea that the ontological argument must fail because it treats existence as a property of objects when it is no such thing. Though the critique is usually aimed at the ontological argument, any philosophical argument seeking to establish that existence is a property or predicate of objects would fall prey to the same claim. However, because the claim is stated as a universal, it is subject to the method of counterexample. If existence is a property of objects, then the Kantian critique alone will not suffice to block the ontological argument’s success. Since there are alternative accounts of existence such that existence is applicable to objects, if the ontological arguer adopted one of these accounts, then the Kantian objection is rebutted.

Other reasons aside from the Kantian critique may result in the ontological argument’s failure – a gap in conceivability and possibility (as in van Inwagen 1998), or that certain modal logics used by ontological arguers are not the appropriate modal logics for metaphysics (as in Salmon 1989), or some other global objection (as in Oppy 1995).

46

But it will not be because existence is not a property of objects. These considerations support premises 2 and 3.

Premise 4 is more difficult to support, but if we examine recent versions of the ontological argument, we note that they are made without any requirement that existence be a property of objects. Malcolm 1960 allows that existence is not a property of objects, but argues that necessary existence could be, and if this is the case then the ontological argument can be made without requiring existence be a property of objects.

Plantinga 1974 and 1977 both contain modal ontological arguments that do not depend on the idea that existence be a property of objects. Maydole 2003 and 2009 also contain modal ontological arguments that do not require that existence be a property of objects.

As before, this is not enough to say that these arguments themselves are valid and sound, as there may be some other substantive objections made to them. But it does show that if these arguments fail, it will not be because they cannot treat existence as a property of objects.

Premise 5 is supported by the foregoing considerations. Whatever the case may be with regard to existence applying to objects or not, the ontological argument fails or succeeds independently of these considerations alone.12

Section V. Conclusion

In this chapter, I have examined two influential and contemporary views that existence is not a predicate or property of objects. I have examined two contemporary

12 An additional argument for the conclusion (5) is given in Oppy 1995 against the Kantian critique and its variations. In Chapter 10 of Oppy 1995, he argues that in order to establish the Kantian critique in any of ways in which it has been thought to debunk the ontological argument, the critic would be required to engage in a serious and entrenched debate about the nature of existence. However, Oppy argues that as long as existence is a predicate of objects in the most minimal and vague sense, they will be able to respond to the Kantian argument. I do not wish to fully explain Oppy’s position here – I note it only to suggest that one major critic of the ontological argument finds the Kantian objection to be significantly undermined as a route of derailing the ontological argument. 47 views that existence is a predicate or property of objects. I have explained an additional view, given by Malcolm, that even if a philosopher were to accept that existence is not a predicate this still might not give sufficient reason to reject the modal ontological argument, for there could be relevant differences between the concepts of “existence” and “necessary existence.” Finally, I have given an argument that, whether or not existence is treated a property of objects, the ontological argument can still succeed, or rather that if the ontological argument fails it will not have anything to do with the difficult question of the nature of existence. In the next chapter, I examine the efficacy of parody arguments.

48

Chapter 3: Parody Objections and AntiGod

Section I. Introduction

Perhaps the earliest objection to Anselm’s ontological argument is found in his contemporary, Gaunilo of Marmoutiers. Gaunilo suggests that one can use Anselm’s strategy to prove the existence of an island “that is more excellent than all other lands...” (Anselm 2007, 104). Of course, there is no such island, and because of this,

Gaunilo thinks that the ontological argument for God fails. In the subsequent years, philosophers have taken to calling such objections parody objections, since they essentially parody the ontological argument by keeping the same form and substituting some other concept for the concept of God (Oppy 1995). In this chapter, I will first motivate the claim that parody arguments are the strongest objections to the ontological argument. Then, I discuss one way of categorizing parody-style counterarguments based on the object that they purport to prove. After distinguishing different parody arguments in this manner, I adopt and examine Oppy’s taxonomy of parodies (Oppy

1995, 162 – 183). Then, I argue that one kind of parody argument, the ontological argument for a necessarily existent, omnipotent, omniscient, perfectly morally evil entity, is a more effective objection than other kinds of parody arguments. I also suggest some dialectical considerations about naming, especially about using the term

“devil” to refer to this kind of entity. I conclude by examining Yujin Nagasawa’s contention that all parody arguments directed against the ontological argument must fail

(Nagasawa 2010). I argue that he fails to prove this point.

Section II. Motivating Parody

49

My argument in this section is brief. In short, I argue that parody objections are the most powerful kinds of objections to the ontological argument. What does it mean to say one objection is “better” or “more powerful” than another? Although this notion is not exactly familiar, I think that the point can be brought out by considering the ways that different objections target a given argument.

Some arguments focus on issues associated with or presumed in an argument.

For example, the Kantian critique discussed in chapter 2 is an example of this kind of objection. The Kantian critique suggests that an important part of the argument rests upon a mistaken understanding of the predicate “exists.” This objection doesn’t out- and-out deny a premise, but rather suggests that some concept or idea or thesis involved in a premise is in need of clarification or correction. Similarly, the conceivability objection focuses on a thesis involved in a premise. These kinds of critique do eventually entail that one should reject a certain premise, at least as it is interpreted by proponents of the argument, but not by outright rejecting the premise, or giving a reason to reject the entirety of a premise. Rather, they focus on something involved or associated with a premise, show that the proponent of the argument is either using this

(whatever is involved or associated with the premise) incorrectly, or that it doesn’t apply as the proponents think, and thus show that the premise is objectionable via this route. Arguing that “existence” is not a perfection is another critique of this form.

Another kind of critique is to give a more general reason to think that a premise ought to be rejected. Arguably, objections of circularity are of this form; they give one a reason to reject the premise outright, not to suggest that it be modified or altered. If one is antecedently convinced that the problem of evil rules out the existence of God,

50 then it could apply as a reason to reject the first premise of any ontological argument, which is that “It is possible that God exists.” Any kind of Moore shift is an example of this kind of rejection-of-a-premise objection.13

Both of these kinds of objections to arguments rely on a disagreement between the proponents and opponents of the argument about some important matter; an inappropriate view of existence, an inappropriate view of the relationship between conceivability and possibility, or a complete rejection of a premise via some kind of

Moore shift. In both of these cases, the opponent of the argument bases the critique in a denial of one or another premise of the argument, and this opens the way to the proponents’ suitably replying by altering or modifying the premises involved.

On the other hand, parody objections proceed in a very different way than the previous two kinds of objections discussed above. Parody objections, if they are successful, act as reductiones ad absurdum. That is, they allow the proponent of the argument all of the kinds of controversial theses and ideas involved in the previous two kinds of critique. In this case, a reductio arguer says “I am not going to deny that conceivability entails possibility, or that existence is a predicate, or any other supposed assumptions of the argument—in short, I will allow you all that you need to make the argument successful.” For another way to put this, the parody arguer argues against the argument on its strongest terms. It grants the proponents of the argument all that they need to make the argument successful, and they then show that if the argument works for x, it can be reproduced to work for y (or in some cases ~x) and this results in a contradiction.

13 By Moore shift I mean the saying “One’s person’s Modus Ponens is another person’s Modus Tollens.” This move was famously used by G.E. Moore in one of his commonsense arguments against skepticism. 51

My view that one objection is “more powerful” than another is based on an idea that we are all familiar with. The problem with a straw-man version of an argument is that it makes the argument weaker than it needs to be. I think something similar, although not fallacious, is going on with the first two kinds of critiques discussed above.

I am not saying that those critiques are examples of the straw-man fallacy, but that like the straw man fallacy, they attempt to weaken the argument and thereby show that it fails. Parody-style objections, since they are examples of the reductio ad absurdum strategy, do not attempt to weaken the argument and show that it fails, but allow the argument to be as strong as possible, and still show that it fails. In this way, parody objections are more powerful than many other kinds of objections to the ontological argument. They allow the argument to be stated in a way that it succeeds, at least initially, and then show that similar styles of argument also succeed, thereby rendering the argument strategy unsound.

Section III. A Two-Pronged Taxonomy of Parody-Style Counterarguments

There are as many styles of parody as there are styles of ontological argument.

In this section, I use a two-level classification system to examine different kinds of parody objections.

III.A The Taxonomy

The first way to analyze parody objections is by examining certain attributes of the objects or entities they are supposed to prove exist. All parody arguments either conclude that a) a physical object or entity of a certain kind exists, or b) an immaterial object or entity of a certain kind exists. Examples of the former are Gaunilo’s island objection, Caterus’ existent lion, and Tooley’s maximal solvent and maximally

52 insoluble substance (in Anselm 2007, Descartes 2008, and Tooley 1981). On the other hand, Henle’s “Nec,” Kane’s LPNs, and the devil of Haight and Haight are all examples of the second kind of entity or object; immaterial, nonphysical entities (in Henle 1961,

Kane 1984, and Haight and Haight 1970). One way, then, to logically map the different kinds of parody is to classify them according to whether they attempt to show the existence of a physical object or entity on one hand, or a nonphysical, immaterial object or entity, on the other.

An immediate question that arises is “Why classify parody arguments this way?

Why is this taxonomy useful?” At the end of this section, I argue that one can dispose of all parodies that attempt to prove the necessary existence of physical objects or entities. If the argument is successful, then by classifying parodies in this way we can eliminate that class of arguments entirely.

Once we have this distinction in place, we can adapt Oppy’s taxonomy of parody objections to fit under either attempts to prove the necessary existence of a physical object or entity, or attempts to prove the necessary existence of immaterial entities. Oppy characterizes parody objections as falling under one of the following kinds: parodies of the form “beings of kind k than which no greater beings of kind k can be conceived,” parodies of the form “most perfect beings of kind k,” parodies of the form “necessarily existent beings of kind k,” parodies of the form “actually existent beings of kind k,” parodies of the form “maximal beings of kind k,” and parodies that

“purport to establish the existence of different kinds of devil,” (Oppy 1995, 162 – 183).

It should be noted that a single kind of parody can fall under more than one of these

53 categories, especially devil parodies. However, in general these categories do distinguish one kind of parody objection from another.

Parodies that rely on the locution “beings of kind k than which no greater beings of kind k can be conceived” attempt to adopt this phrase from Anselm’s argument and then substitute a concept other than God for the subject of the initial possibility premise.

An obvious example of this is the island parody. It follows the form of the ontological argument except that instead of “a being than which none greater can be conceived” the entity involved in the argument is “an island than which none greater can be conceived.” If the argument works in the case of God, then the parodist suggests it works as well for an island, or for any other kind of thing. There is nothing special about using an island in this way, as the parody admits of generalization. Presumably

“a tree than which none greater can be conceived,” “a basketball player than which none greater can be conceived,” and “a motorcycle than which none greater can be conceived” would prove the same point that the island does. One need not limit oneself to physical entities in this kind of critique, although the examples often discussed in relation to this critique frequently are physical entities. Since this kind of parody involves the notion of greatness, then questions about greatness or various definitions of greatness will need to be answered or settled to not only see if the original argument works, but also to see if these parodies can apply at all. We will come back to the notion of greatness and questions about it in a later chapter.

Parodies of the form “most perfect beings of kind k” are quite similar to the first kind of parody discussed in the preceding paragraph. The primary difference, at least in the eyes of some philosophers, is the distinction between greatness and perfection.

54

Robert Brecher, for example, argues that Anselm did not use “greatness” and

“perfection” interchangeably, but that considering God’s greatness is different than just considering God’s perfection (Brecher 1974). If this is true of Anselm’s argument, then there is no reason that contemporary arguments that utilize the terms “perfection” and or “greatness” cannot hold the same thesis. Whether or not there is a significant difference between perfection and greatness, since this kind of parody uses the term

“perfection,” then just like the parody discussed in the preceding paragraph, any questions about perfection or various definitions of perfection will need to be answered or settled to see if both the original argument works and to see if the parody-style critiques can even get off the ground.

Parodies of the form “necessarily existent beings of kind k” are, in some ways, a bit more artificial than the two kinds of parodies just sketched above. It might be thought that God’s essence implies necessary existence, but how does that show the viability of just any old concept constructed to contain the property “necessary existence?” For a theist, such constructions will seem unnecessarily artificial and ad hoc, but the atheist or agnostic may have a different view on just how organic the concept of “God” is in the first place. If the concept of “God” and the concept of “a being that, among its properties, has necessary existence” are either equally controversial or equally ad hoc, then the parodist can proceed with the argument. Of course, it may be the case that the parodist needs to argue for this parity, and it need not be the case that every theist must agree that the concepts are on par in this respect.

What we can show initially is just that there is no logical reason to rule out “necessarily existent beings of kind k” at the outset. Henle and Kane both adopt this parody

55 strategy, although how they go about making their respective parodies is a bit different

(Henle 1961 and Kane 1984). A crucial question to ask for this kind of parody (and for this kind of ontological argument) is “What kinds of things can even possibly possess necessary existence?” The answer to this question is not at all obvious but proponents of both parodies and ontological arguments that use this feature need to have an account of it. It will be in the interest of proponents of the ontological argument to provide a properly motivated account of necessary existence that involves a principle that shows that necessary existence can only be properly predicated of God. It will be of equal interest to parodists to develop an account of necessary existence that shows that it can, indeed, be properly predicated of at least one other entity than God. One thing that will be true of any account of necessary existence is the relationship between necessary existence and possible worlds; something exists necessarily if and only if it exists in all possible worlds.

Parodies of the form “actually existent beings of kind k” are, as those posited necessary beings given in the previous section, more artificial looking than some of the other parody objections discussed. The parody of this form suggests that, given

Descartes ontological argument (and more broadly, any Cartesian style ontological argument) one can prove the existence of any kind of thing, or one can prove the existence of any kind of thing that only exists contingently, and therefore could have failed to exist (Oppy 1995, 173 – 179). The second form of this critique is found in

Caterus’ reply to Descartes. Caterus supposes that, in the same way Descartes moves from examining the concept of God to proving that God exists, one can contemplate the concept “existent lion” and move from the contemplation of this concept to the

56 existence of this concept. Objectors of this sort will insist that just as some kinds of ontological argument for the existence of God smuggle the concept “actual existence” into the concept “God,” one can include the concept “actual existence” into any kind of thing, including contingent objects, and thereby prove their existence. The problem here is that one can prove, a priori, the existence of some contingent thing—and presumably, we should need to do some investigation before we settle the question of the existence of something that either could or could fail to exist. If this parallels the ontological argument, then it seems that the argument strategy is simply reifying the concept of God, and is not to be given serious intellectual consideration. The first form of this critique as discussed above can be found in the work of Rowe. After analyzing and laying out a version of Anselm’s argument, Rowe writes

We shall say that the term ‘magician’ may be applied both to Houdini and Merlin, even though the former existed whereas the latter did not. Noting that our friend has used ‘existing’ as part of his definition of ‘God’, suppose we agree with him that we can define a word in any way we please, and, accordingly, introduce the following words with the following definitions: ‘a magican’ is defined as an existing magician, ‘a magico’ is defined as a non-existing magician. Here we have introduced two words and used ‘existing’ or ‘nonexisting’ in their definitions. Now something of interest follows from the fact that ‘existing’ is part of our definition of ‘a magican’. For while it’s true that Merlin was a magician it isn’t true that Merlin was a magican. And something of interest follows from our including ‘non-existing’ in the definition of ‘a magico.’ For while it’s true that Houdini was a magician it isn’t true that Houdini was a magico. Houdini was a magician and a magican, but not a magico; whereas, Merlin was a magician and a magico, but not a magican. What we have just seen is that introducing ‘existing’ or ‘nonexisting’ into the definition of a concept has a very important implication. If we introduce ‘existing’ into the definition of a concept it follows that no non-existing thing can exemplify that concept. And if we introduce ‘non-existing’ into the definition of a concept it follows that no existing thing can exemplify that concept. No nonexisting thing can be a magican, and no existing thing can be a magico. (Rowe 1976, 427 – 428)

57

Rowe argues that if we can form the concept of God as including “existence” then we can do the same for any other concept. It need not be a magican, an “existing magician,” but can be generalized, as Oppy and Rowe note. For any possible object, simply create a new or altered concept that is the concept of that possible object plus

“actual existence” or just “existence.” Rowe claims that this shows that in the case of the ontological argument, just as in the case of a “magican,” only that “God” or

“magican” cannot be identified with any non-existent object, not that “God” or a

“magican” must be identified with some existent object (Rowe 1976, 428 – 429). I take it that this kind of parody is supposed to show that since the theist (and anyone in their right mind) will not countenance such a move for magican or a similar concept, then they should also reject the strategy as it applies to God. This form of the parody comes to the same conclusion as the other form of this parody, which is that this argument strategy is simply reifying something into existence, and this is a serious flaw.

Parodies of the form “maximal beings of kind k” are directed at Plantingian ontological arguments. Recall that Plantinga’s ontological argument introduces the concept of maximal greatness and maximal excellence. He defines maximal excellence as “omniscience, omnipotence, and moral perfection” and maximal greatness as “having the property of maximal excellence in every possible world,” (Plantinga 1974, 218).

Although Oppy, following Tooley, states that “x is maximally p iff x exists in all possible worlds, and is p in every world,” this seems to be a slight misunderstanding of

Plantinga (Oppy 1995, 179, and Tooley 1981, 424). It is true that the property

“maximal greatness” does follow the form Oppy suggests, but the property “maximal excellence” does not. This is a minor point, as at least some kinds of maximality do

58 entail necessary existence, but it is worth getting clear on definitions used in a major philosophical argument. We can assume for the rest of this section that the sense of maximality in question is the one Plantinga applies to greatness, and is as Oppy suggests. Tooley gives an ingenious argument for two maximal objects, both of which are necessarily existent, and the existence of both in the same world creates a logical contradiction. Consider the property “maximal universal solvent” where that means “x is a maximal universal solvent iff x exists in every world and is a universal solvent in every world, where something is a universal solvent in a given world if and only if it is capable of dissolving anything in that world.” (Tooley 1981, 424). Let us also consider the property “maximally insoluble” where that means “x is maximally insoluble if and only if x exists in every world and is insoluble in every world,” (Tooley 1981, 424).

Tooley argues that we can parody Plantinga’s argument by using the same argument form, but by arguing for the instantiation of the property “maximal universal solvent” and by arguing for the instantiation of the property “maximally insoluble” instead of the property of “maximal greatness” in either case. But surely this is incorrect; if something in a world is capable of dissolving anything in that world, there cannot be a maximally insoluble object. Alternately, if something in a world is maximally insoluble and, by definition, cannot be dissolved by anything in that world, then nothing in that world can have the property of being a maximal universal solvent. If a modified version of these objects were shown to exist in some worlds but not others, we would hardly have a problem. If “universal solvent” is instantiated in a given world, then

“insoluble” cannot be, and vice versa. The problem is caused by the “maximal” predicate. If maximal greatness is possibly instantiated, then it exists in every possible

59 world, and the same is true of maximal universal solvents and maximally insoluble objects. Of course, we may want to reject the possibility premise in the modal arguments for maximal universal solvents and maximally insoluble objects, but this same attitude can be taken by the atheist or agnostic as regards the possibility premise in the modal argument for God. What the theist needs is a principled reason to reject the possibility premises involved in Tooley’s parody arguments, but that does not rule out the possibility premise of the modal ontological argument. The proponent of the parody will want to provide grounds to reject this distinction. Here, then, is a straightforward case of a reductio aimed at the argument style evinced by modal ontological arguments. If we can use this argument style to prove the necessary existence of two objects that are not compossible, something has gone terribly wrong.

The final category of parody arguments is arguments that attempt to establish the existence of some kind of devil or another.14 This kind of argument is often of the form of one of the previous style of parodies we have examined above. This category of parody arguments is better understood, not as classifying a different form of parody, but as classifying certain parodies by reference to the type of object they are supposed to prove. There are many examples of arguments that mirror the ontological argument but substitute an omniscient, omnipotent, perfectly morally evil entity in the place of an omniscient, omnipotent, perfectly morally good entity; a number of philosophers have all given ontological parody arguments for the existence of some kind of devil or similar entity (Chambers 2000, Cock 1918, Gale 1993, Haight and Haight 1970, Haight

1974, Johnson 1965, Millican 1989, Millican 2004, Millican 2007, Richman 1958,

14 While I use the term “devil” throughout section III, I argue in section IV that usage of this term is problematic. 60

Richman 1960, and Wilbanks 1973). I will compare this kind of parody to other styles of parody at length in section IV, so I will not do so here. The common feature running through this category of parody arguments is that they all attempt to show that a devil, or some kind of omnipotent, omniscient, perfectly evil entity, can be shown to exist in the same way that the ontological argument proves the existence of God. Since the proponents of these parodies, unless they are Zoroastrians or Manicheans, do not endorse the existence of such a being, the purpose of these parodies is to show that the ontological argument is not a sound argument strategy. This proceeds via the familiar reductio ad absurdum strategy. This could be done at least one of two ways; first, by showing that the necessary existence of two omnipotent deities of any kind implies a logical contradiction, and second, by showing that the necessary existence of two omnipotent, omniscient entities with exactly opposite moral compasses implies a logical contradiction. Either way, the original argument conjoined with the parody for a devil results in a contradiction, and therefore the ontological argument fails.

With all of this in mind, we can give a taxonomic tree of parody-style arguments. In order to do so, I have shortened the names of the forms of the parodies as follows: parodies of the form “beings of kind k than which no greater beings of kind k can be conceived,” I will label “Anselmian”; parodies of the form “most perfect beings of kind k,” I will label “perfect,” parodies of the form “necessarily existent beings of kind k,” I will label “necessary,” parodies of the form “actually existent beings of kind k,” I will label “actual,” parodies of the form “maximal beings of kind k,” I will label

“maximal,” and parodies that “purport to establish the existence of different kinds of devil,” I will label “devil.”

61

III.A Disregarding Physical Object Parodies

Earlier I answered the questions “Why classify parody arguments this way?

Why is this taxonomy useful?” by promising to give an argument that will allow us to dispose of any counterargument that attempts to deduce the existence of a physical object via an ontological parody. I do so here.

The key premise in this argument is alternately “No physical object is necessary” or “No physical object exists necessarily” or “Every physical object is contingent,” or “Every physical object contingently exists.”15 If this premise holds true, then we can make the following argument.

1. Every physical object is contingent. 2. Ontological arguments, including parody-style ontological arguments, are a priori arguments. 3. Some set of parody arguments either show the a) actual existence of a physical object, or the b) necessary existence of a physical object. 4. But b is impossible, from 1. 5. An a priori argument cannot deduce the existence of a contingent object, since for any contingent object, the world could be such that it does not exist. 6. Since finding out whether a contingent object exists requires us to go out and examine the world in which we live, showing the existence of such an object requires a posteriori methods. 7. So a parody argument for a requires an a posteriori method of argument. 8. So a parody argument for a is not an a priori argument. 9. Therefore, it is impossible to parody the ontological argument by giving a parody argument for b since there is no such thing as a necessary physical object, and it is impossible to give a parody argument for a since it will not be an a priori argument, which is ruled out by 2. 10. So any ontological parody argument for a physical object of any kind fails.

15 While I argue for this later, for another defense of such a principle, especially in connection with certain kinds of parodies, see Mann 1976. 62

Since premise 1 does a lot of the heavy lifting in the argument just given, it requires defense.

What is a physical object? I take it that a physical object is an object composed of the entities given in various levels of physics. For any physical object, we can imagine that it not exist; indeed, we know that there are specific combinations of matter and energy that allow for the fading out of existence for such an object. If I am a purely physical object, then there are many conditions under which I would cease existing, such as being in a room completely devoid of oxygen, or being immersed in a stream of lava. Further, this is true of any physical object. Coconuts, islands, books, stars, galaxies, molecules, and particles all have conditions that are required for their continued existence and all have conditions under which they cease to exist, or are destroyed. This point deserves emphasis. A coconut can be smashed into smithereens, islands can be eradicated by earthquakes or a meteor impact, books can be burned or shredded, stars go through various stages of life culminating in their eventual destruction, and so on, and so on. If all physical objects are composed of the elementary particles found in physics (governed by appropriate laws and subject to appropriate forces) and these elementary particles themselves can be destroyed or can fail to exist, then any object composed of these particles can be destroyed, or fail to exist. This point is especially salient when we realize that an object can be destroyed in at least two ways; first, by destroying all of the articles that compose it, and second, by rearranging the elementary particles in suitable ways. If, through some amazing, natural law-governed process, all of the particles that compose me change in such a way that I turn into a viscous slime, or a large clock, then it seems correct to say that I am no

63 longer. Since every physical object we can conceive of can be thought to not exist or can be shown to have conditions under which they either continue to or fail to exist, then talking about a necessary physical object is simply nonsense.

Suppose, though, that my interlocutor claims I have done the argument an injustice. “I can conceive of a necessary fork; what I mean by this is that in every possible world, this fork exists. What is contradictory about such an imagining?” In this case, we need to ask if the fork is a purely physical object, to which my interlocutor surely agrees. Then, we need to ask, as I did at the beginning of the preceding paragraph, “What is a physical object?” If the answer is the one I gave (and I can think of no reason my opponent would deny such an answer) then it looks like there could be conditions under which the fork ceases to exist; it is melted down, or chopped into a fine dust, or some other set of conditions under which it ceases to exist. How can my interlocutor reply? To simply say that these conditions are impossible is to say that either a) this fork is not a purely physical object, and thus has some ethereal quality or qualities that preserve its existence through the changes described above, or b) that it is a physical object composed of physical entities, by which we mean something very, very different than the physical objects and the entities that compose them in our world.

The first horn of this dilemma just rejects that it is a purely physical object, which is the only thing I am interested in considering in terms of these parodies. The second horn may seem more promising, but now we must ask for an account of this wildly different physics by which something like a fork can be necessarily existent. If it is too divergent, as I think it must be, then we are no longer talking about physical objects properly conceived, but physical’ objects, or physical’’ objects that have only the

64 scantest relationship to our physical objects. In summary, there is no good reason for thinking that any physical object can be a candidate for necessity, and there are good reasons to think that for every physical object, we can give an account for what is needed for its continued existence and the conditions under which it would cease to exist, thus making premise 1 justified.

The second part of the argument, showing that parodies that attempt to prove the actual existence of physical objects, also requires some explanation. A hallmark of ontological arguments (including ontological parodies) is that they are a priori. This is a complex and often murky philosophical notion, and I will defend no particular account of it here.16 Instead, I rely on the judgment of myself and the reader that a priori is understood as a proof that proceeds from purely conceptual and logical grounds, without certain kinds of reference to empirical investigation into the world. If an object is contingent, then it could fail to be, and only an examination of our world will show that the object does not exist, or that it does in fact exist. If we use these grounds, then our parody is no longer a parody of the ontological argument, but rather an a posteriori argument that some physical object exists. So a parody cannot prove a priori the actual existence of physical (and from premise 1, contingent) objects.

If the argument in this section is correct, then we can leave unexamined any parody for the existence, actual or necessary, of physical objects. Gaunilo’s island,

Tooley’s maximal universal solvent and maximally insoluble substance, and Caterus’ lion can all safely be rejected by the proponent of the ontological argument.

Section IV. Angra Mainyu, the Devil, and AntiGod

16 For a good overview of different conceptions of “a priori” and the different ways the term has historically been used, see Moser 1987. 65

In this section, I first discuss some concerns about the usage of the term “devil” in these parodies. Then, I argue that of all parodies that deduce the existence of immaterial objects, the class of parodies that attempts to prove the existence of an omnipotent, omniscient, perfectly morally evil entity is a stronger kind of parody than other parodies that attempt to prove the existence of immaterial objects. Recall that since parodies of this kind can be of different forms, for example “a being than which no worse can be conceived,” “a most perfect demon/devil/etc.,” “that maximal malevolence be instantiated,” the argument in the second subsection of this section will argue against other kinds of immaterial entities such as Henle’s Nec or Kane’s LPNs.

Finally, in the last sub-section of this section I briefly examine the argument given by

Haight and Haight 1970 since this is the most direct form of the argument. Interested readers are referred to Millican 2004 and Chambers 2000 for more recent versions of this kind of parody.17

IV.A Terminology and Connotation

Although it is common to call these kinds of parodies arguments for the existence of a devil or devils, I think that this can impede clear consideration of the topic. Perhaps little hangs on this point, but even if the reader is not as convinced as I am of the seriousness of this problem, I hope she will at least take this discussion under consideration.

17 I developed a version of the ontological parody argument for the existence of AntiGod independently of the authors I’ve mentioned in the summer of 2005 at the Colorado Summer Seminar in Philosophy at the University of Colorado, Boulder. Since I think that the central insights in my version of the parody argument are essentially the same as those contained in other parodies given by Haight and Haight 1970, Chambers 2000, and Millican 2004, among others, I will not produce another version of this objection here. 66

First, the devil in the Christian tradition is not omnipotent, omniscient, and perfectly morally evil. As a fallen angel, the entity is neither omnipotent nor omniscient. While this seems like an obvious point, I think arguments for the existence of a devil hold a connotation that the being in question is substantially lesser than God, even if this is not formally the case.

Second, strong emotional responses are evoked by usage of the term “devil.”

For many Christians, the devil is not a humorous horned figure from pop culture wearing red tights and exhibiting a sinister mustache, but a significant contributing factor to the fallen state of nature and a figure that is constantly tempting humans with the goal of corrupting their immortal souls. Are we, as professional philosophers, immune to such an emotive association? The answer is almost certainly no. I do not mean to suggest that one must view the devil in this way if one is a Christian, only that the contingent facts of the matter are such that many people in fact do. I also note that this point is true of the term “God.” Many atheists have very strong negative emotive associations with the term “God,” and I think this can explain some of the psychological hesitance to endorse something like the ontological argument. It is obvious that in the case of the ontological argument, proponents will have no desire to change the name of the entity under consideration in order to level the emotional playing field between atheists and theists. In the case of devil parodies, though, nothing save our entrenchment in natural linguistic practices constitutes a reason to continue using a term that has these connotations.

I am suggesting that because of the historical conceptual web surrounding the term “devil” both in a sense that there is a connotation of inequality, and in the sense

67 that there is a negative evaluative connotation, we should replace the term with something less controversial that captures the meaning we desire to capture. In

Manichean ontology, there is a King of Darkness that is the opposite figure of the

Father of Greatness, but these conceptions rest on a gnostic view which is itself the subject of much emotive association. In addition, the phrase is simply not short enough to be useful in this context. Angra Mainyu, the evil entity or principle that opposes

Ahura Mazda in Zoroastrian religion is closer to this meaning, although as Angra

Mainyu is not exactly the equal of Ahura Mazda, and since only in some traditions is

Angra Mainyu seen as an entity rather than a principle, this will probably not do either.

Millican’s “AntiGod” is probably the least controversial artificial name for such a being, although adding the prefix “anti-“ to another thing may strike one as simply contrarian or childish. In short, we have to choose between using a name we are all familiar with, that has certain historical baggage, but that people will likely take seriously, and a name that is stipulated, thereby free from historical baggage, but that may appear to be too playful or artificial to be taken very seriously. Since the aim of parodies of this kind is to make the theist take seriously this parallel argument and to contemplate it with as open a mind as possible, it is in the interest of the parodist to use a term that is not too playful or artificial to be taken seriously, but that also avoids the connotations that may subconsciously influence the appraisal of such an argument.

I confess that I would prefer to use an artificial term, and be accused of jocularity in the face of a serious argument, than use an existent historical term with unhelpful connotations. Although Millican’s “AntiGod” is a bit more playful than I would prefer, I cannot think of another short, pithy alternative. Throughout the rest of

68 the section and the rest of the dissertation I refer to devil parodies as AntiGod parodies or as parodies for the existence of AntiGod, but the reader should note that I do not mean this in a way that is playful or cavalier. I take this objection, as I hope the reader does, with the utmost seriousness and sincerity. I also note that the taxonomic tree developed above would be slightly different in terms of the appropriate labels.

IV.B The Supremacy of AntiGod Parodies

Since, in section III.B I argued against the need to consider parodies that deduce the existence, of any kind, of physical objects, then that leaves only the parodies that attempt to prove the existence of immaterial objects as the subject of our consideration as possibly successful parodies. Since I am primarily concerned in this project with contemporary modal ontological arguments, then I leave aside parodies of the form

“actually existent beings of kind k” as applied to immaterial entities.

My argument in this section has less to do with the forms of various parodies for the existence of immaterial objects, and more to do with the kind of immaterial object proven to exist. It can be summarized as follows.

1. For any immaterial object purportedly proven by an ontological parody, it is either AntiGod, or it is something else. 2. AntiGod has advantages x, y, and z (to be spelled out below) over other immaterial parody objects. 3. We ought to present the parody objection in the strongest form possible. 4. Therefore, if we argue against the ontological argument by parody, we ought to argue for the parody of AntiGod.

We now move to consideration of AntiGod as compared to Henle’s Nec brethren, and

Kane’s LPNs. These are, aside from AntiGod parodies, the best kind of immaterial object parody to modal ontological arguments. If I show that AntiGod has advantages

69 over these kinds of immaterial objects, it can be concluded that any other immaterial object parodies also fail to be as good as or better of an objection than AntiGod.

In order to compare AntiGod to other kinds of necessarily existent immaterial beings proved via parody, it will be important to lay out the features of both Henle’s

Nec brethren and Kane’s LPNs. Since both Henle and Kane utilize an almost identical strategy in enumerating the different necessary beings that each think can be proven with a parody argument, I will examine each view briefly before considering them in tandem.

Henle argues that, in much the same way Malcolm argues for a modal ontological argument for the existence of God, one can adapt the argument to prove the necessary existence of other beings. He introduces Nec, NEc, and NEC. Henle writes

Let us designate by ‘Nec’ a certain being who has necessary existence but who is otherwise less remarkable. He has a certain amount of knowledge, though nothing extraordinary, and certain power, although he is unable to cause motion. As a necessary being, of course, Nec’s nonexistence is inconceivable and he does not depend on anything… Even though Nec’s existence precludes that of an omnipotent being, it is perfectly compatible with that of other beings having necessary existence. Nec has a big brother, NEc, who also exists necessarily but who can cause uniform rectilinear motion and is a little wiser than Nec. There is another brother, NEC, who can cause acceleration, and only typographic inadequacies prevent me from enumerating a spate of others. (Henle 1961, 102 – 103)

Similarly, Kane argues that accepting the appropriate principles of modal logic and modal systems (such as S5, or perhaps Brouwer) that the modal ontological argument relies on, one can use these same principles and systems to deduce the necessary existence of a class of beings, called “less-than-perfect necessary beings,” or

LPNs for short. He writes

70

Let us consider the possible existence of what might be called ‘less-than- perfect necessary beings’ (LPN’s). I mean beings whose essence or definition is such that, if they exist, then necessarily they exist, but which lack some other attributes of perfection, e.g., they are less than omniscient or not omnibenevolent. Are LPN’s possible in the broadly logical sense? Is it self-contradictory to say that a being has necessary existence but is not all good or all knowing? If the answer is negative, then there might be an indefinite number of such beings, e.g., one which commits at least one act in every possible world, one whose acts are always evil (a perfect devil? An evil god?), one which knows certain things, but not others, and so on. Furthermore, it would be possible to use the principles of the modal OA [ontological argument] to derive the actual existence of any such possible LPN. If the B-principle holds, and if an LPN exists in some possible world relative to the actual world, then it exists in the actual world. Moreover, the LPN would necessarily exist in the actual world because it is such that, in every world in which it exists, it necessarily exists. (Kane 1984, 344 – 345)

Now that we have laid out the relevant concepts in the authors’ own respective words, we can examine the similarity in the approach of both kinds of parody arguments for the existence of necessary immaterial entities.

Both Henle and Kane’s parody arguments have the same structure. Take the modal ontological argument, and alter slightly certain properties involved in the concept of God, keep necessary existence as a constant, and change the properties such as omnipotence, omniscience, and omnibenevolence or moral perfection. Tinker with various combinations of these concepts to produce fleshed out concepts of Nec, an

LPN, and their ilk. Oppy has an excellent discussion of how to produce a large number of these kinds of entities, which I quote in full.

It seems that we can generate infinite numbers of LPNs. For convenience, the following discussion will presuppose that omniscience, omnipotence, and omnibenevolence are the only relevant attributes of God. It seems to me that this presupposition is harmless in the present context. First, suppose that we can quantify benevolence using real number percentages. If we make the assumption that moral properties are not necessarily linked to knowledge and power—that is, any kind of moral property is compatible with omniscience and omnipotence—then,

71

for each 0 < r < 100, there is an LPN that is omniscient, omnipotent, and possess benevolence to degree r. Of course, God possesses benevolence to degree 100—that is, in the form of omnibenevolence. And if we deny the assumption that moral properties are not necessarily linked to knowledge and power—on, say, the grounds (i) that there is, properly speaking, moral knowledge, and (ii) that moral knowledge is intrinsically (or internally, or necessarily) linked to moral motivation—we can still maintain that there is an LPN that is all but omniscient, all but omnipotent, and that possesses benevolence to degree r. Second, we can suppose that there is a hierarchy Oi of omniscient, or all but omniscient (and omnibenevolent, or all but omnibenevolent) LPNs that are also all but omnipotent in the following way: For each i, Oi is limited only by Oj, for all j > i. Thus, for example, O2 can do things O1 cannot do, but not vice versa. When it comes to a competition between O1 and O2 in which there is a winner, the winner is always O2. This hierarchy can extend into the transfinite, if we wish. Third, we can suppose that there is a hierarchy of all but omnipotent (and omnibenevolent, or all but omnibenevolent) LPNs that are also all but omniscient in the following way: For each i, Oi knows everything about things outside the hierarchy and also everything about beings below it in the hierarchy—that is, about Oj, for all j < i—but does not know some things about itself and beings above it in the hierarchy—that is, about Oj, for all j ≥ i. That is, these beings are omniscient about worldly matters of fact, but ignorant about some things to do with themselves. In this discussion, I have supposed that omniscience is a consequence of omnipotence. This might be denied; but, since an omnipotent being ought to be able to answer every question correctly, it does seem that only an omniscient being can be omnipotent. (Oppy 1995, 171 – 172)

I think Oppy’s construction of these various LPNs, and his specifications that allow us to create more and more of them, gives us a good idea how to adopt this procedure.

Notice also that Nec, NEc, and NEC can be categorized under the rubric given above.

The primary similarity, then, between Henle’s Nec, NEc, and NEC and Kane’s

LPNs is that in both cases, the proponent of the ontological parody argument develops the concepts of these immaterial entities by first taking the concept of God and then tinkering with various degrees of power, knowledge, or some moral quality given in that concept to produce a host of immaterial beings with varying degrees of potency, epistemic acumen, and good or evil.

72

I believe that this similarity is the point of attack for someone who, like this author, thinks that AntiGod objections are superior to objections that prove the necessary existence of Nec and LPNs. I do not mean that one cannot use the procedure described above to develop a conception of AntiGod, but only that one need not use such a procedure. I contend below that the ontological parody argument for AntiGod is superior to arguments for LPNs for two reasons; that they are more natural, and less artificial, than proofs for Nec and other LPNs and that they have a greater similarity to the ontological argument for God than do arguments for Nec and other LPNs.

First, the concept of an omnipotent, omniscient, perfectly evil being is a more natural concept than the concept of Nec or other LPNs. I am not sure how to give an exact definition of the ways I am using natural and artificial, but some examples may help to bring out this distinction. Natural concepts include: blue, lion, human, and

(arguably) God and AntiGod (“the devil” is a better historical name for this natural concept, but, as stated above, I do not use such a label here). Artificial concepts, in contrast to the list just provided, include: grue, chimera, Swampman, and (arguably)

McEar, Nec and LPNs. Natural concepts have the two features, then, having a place in the everyday conceptual schemes of human beings, and they do not involve ad hoc stipulations concerning their attributes. Artificial concepts do not usually have a role in the lives of everyday persons, and often suffer from some ad hoc stipulating about their attributes. Why would we think AntiGod is a natural and not artificial concept?

Humans have, quite often, viewed the world as a place subject to the grip of both benevolent and malevolent forces. Further, in the case of many historical mythologies, there is a set of entities that are good and affect and influence the world, and a set of

73 entities that are evil and affect the world. This pattern occurs in Manichean,

Zoroastrian, Greek, and Norse mythologies, as well as in the contemporary religious traditions of (some kinds) of Christianity, Hinduism, and Islam. On the other hand, grue, chimera, Swampman, and McEar, Nec, and other LPNs are examples of highly tinkered concepts that are mostly developed in order to relate to specific conceptual schemes or thought experiments. This does not mean that these concepts ought to be rejected out of hand, nor does this mean that they cannot do important philosophical work. But I argue below that in certain cases, natural concepts have a dialectical advantage, and so if we have two arguments available against a position, and one involves natural concepts while the other involves artificial concepts, the former is preferable.

Even if the foregoing is granted, and it needs further argument, why would natural concepts be thought to have any advantage? I think natural concepts of the kind sketched above are in a better position than artificial concepts for two reasons. First, I think they are simply more dialectically efficacious because they are concepts that we are all, in some hazy, inchoate way, familiar with. Thus, it is psychologically easier for us to conceive of such beings and to take them seriously as articles of philosophical contemplation. Second, I think the highly specified and ad hoc way that many artificial concepts are produced opens them up to a host of objections that do not straightforwardly apply to natural concepts. Arguments about how to understand

Swampman in the light of evolutionary, teleological accounts of words and concepts do not arise for the concept of human. Concerns about whether grue is really a kind of property, and how it should be understood, do not arise in the consideration of the color

74 blue.18 And the same holds true for McEar, Nec, and LPNs as compared to God and

AntiGod.

AntiGod parodies also have an advantage of similarity or parity with not only the ontological argument, but also with the more general concept of God that Nec and

LPNs lack. First, they have greater parity with the ontological argument because while

Nec and LPNs can only be a part of parody of the form “necessarily existent being of kind k” AntiGod parodies can, much like ontological arguments, be formulated in different ways as e.g., “a being than which none worse can be conceived” “a most perfect evil being” “that maximal malevolence be instantiated” “actually existent

AntiGod” and “necessarily existent AntiGod.” Since AntiGod parodies are classified by the subject they purport to prove, and not of a certain form of parody, they can be adapted, with greater or lesser levels of success, to every form of parody given, whereas

Nec and LPNs cannot. Just as the ontological argument for God may be formulated in various ways, so too with the ontological parody argument for AntiGod.

Second, the concept of AntiGod is a fairly natural mirror to the concept of God, as both hold omniscience, omnipotence, and necessary existence, but have diametrically morally opposed plans for the world. Because of this similarity, we can mirror many theological issues with AntiGod in ways that make little sense with regard to Nec and

LPNs. Here are a host of arguments related to God that can be paralleled cleanly by

AntiGod parodies, but that do not obviously or immediately apply to Nec and LPNs; the Cosmological Argument, the Teleological Argument, arguments from religious experience (both cosmological and teleological arguments can show the existence of

18 Grue refers to a mixed predicate used in Goodman 1973 in order to illustrate the new riddle of induction. Swampman, a molecule-for-molecule clone of Davidson created by a freak lightning strike, is first mentioned in Davidson 1987 in order to illustrate an issue regarding consciousness and meaning. 75

God as well as the existence of AntiGod, and presumably one can have an experience, not of unity and transcendence, but of discord and horror), The Problem of Evil (and related Problem of Good), concerns about whether God must create the best world

(must AntiGod create the worst world?), free will’s value in the context of creation (if freely chosen love of and relationship with God is the more significant, then isn’t freely choosing to do evil and harm also more morally significant?), free will as a defense against the problem of Evil and/or Good (if free will is valuable as suggested above, then there is both a reason for God and AntiGod to allow it to be a feature of humanity), and others.

Having shown the superiority of AntiGod parodies compared to other parodies that prove the existence of immaterial objects, we now turn to presentation of the

AntiGod ontological parody.

IV.C Haight and Haight

In this subsection, I present the arguments of Haight and Haight for an AntiGod

(or as Haight and Haight refer to the entity, the devil).

The argument of Haight and Haight is as follows.

Anselm’s first argument, roughly, is as follows: 1. I have a concept of something ‘than which nothing greater can be conceived. 2. If that ‘something’ did not actually, or in fact, exist, it would not be ‘that than which none greater can be conceived,’ for something could always be conceived to be greater, viz., something that actually exists. 3. This ‘greatest something’ is, by logical equivalence, or definition, ‘God.’ 4. God exists. An ontological argument for the devil, by analogue of reason, goes as follows: 1. I have a concept of something than which nothing worse can be conceived.

76

2. If that ‘something’ did not actually, or in fact, exist, it would not be ‘that than which nothing worse could be conceived,’ because something could always be conceived to be much worse, viz., something that actually exists. 3. This ‘greatest something’ we shall call the Devil. 4. The Devil exists. This second ontological argument, by parity of reasoning with the first, seems sound, if indeed, the first is. Is it not conceivable that not only do we have an idea of something that is the worst possible thing, but that it would have to exist if it truly were the worst possible thing? Hence, the very possibility of the Devil implies his actuality, just as the very possibility of God implies his existence. The logic is the same, in both cases: a devil would not be the Devil unless he existed and was therefore the most awful thing, just as a god would not be God unless he existed was therefore the greatest thing. (Haight and Haight, 1970, 218 – 219)

Haight and Haight’s version of the argument is the most intuitive formulation of the AntiGod parody. Haight and Haight make an important point when they say “Is it not conceivable that not only do we have an idea of something that is the worst possible thing, but that it would have to exist if it truly were the worst possible thing?” (Haight and Haight 1970, 219). The idea runs parallel to Anselm’s argument, that if we imagine the most perfect good being, a being with all of the same characteristics would be greater if it existed than not, presumably at least for reasons of efficacy. Isn’t the same true of the most perfect evil being? It could presumably be a more significant evil if it were existent and efficacious rather than not. Surely I would prefer to be the object of hatred and stalking of a nonexistent perfectly evil being than one that hates and despises me but that is also existent. Haight and Haight also evince a point I made above about

AntiGod parodies, which is that they are capable of being employed in any form of parody that mirrors the suitable formulation of the ontological argument, whether it is based on Anselm or Plantinga or Gödel’s reasoning.

Section V. Must Parody Arguments Fail?

77

In the final section of this chapter, I examine Yujin Nagasawa’s persuasive and forcefully argued “The Ontological Argument and the Devil,” (Nagasawa 2010).

Nagasawa not only claims that both the AntiGod objection of Peter Millican and the extreme-no-devil corollary of Timothy Chambers fail to appropriately parody the ontological argument, but also that any parody whatsoever is required to fail as compared to the ontological argument. He argues that all parody objections fail for at least one of the following two reasons: that parody arguments fail to be structurally parallel to the ontological argument, or that parody arguments fail to be dialectically parallel to the ontological argument (Nagasawa 2010, 91). To illustrate these cases, I will present Nagasawa’s arguments against both Millican’s AntiGod, Nagasawa’s own devil argument, and the island objection of Gaunilo.

V.A The Global Challenge to Parody

Prior to going on the offensive, Nagasawa expresses great appreciation for the power of parodies. He notes that it is often the case that the parody objection

(presumably this means any kind of parody) is seen as just an odd consequence of the argument, or a “second-class supplementary objection,” (Nagasawa 2010, 77). He writes that “Few critics regard it as something that could constitute a serious objection to the ontological argument,” (Nagasawa 2010, 77). However, Nagasawa thinks that parody objections deserve to be appraised at a higher value than this. The two primary strengths of parody objections that he considers are also strengths that I have discussed earlier. First, he regards the fact that parody objections accept all of the controversial metaphysical assumptions of the ontological argument as a virtue. This is in line with what I argued earlier was a strength of parody objections, namely that parody objections

78 face the argument in its strongest form, without getting involved over queries about the nature of existence or the relationship between conceivability and possibility.

Second, he notes (rightly) that parody objections are to be taken seriously because they can be modified and adapted to a wide variety of versions of the ontological argument. Many other objections to the ontological argument are only objections to specific kinds of ontological argument. Thus, Norman Malcolm argues that the version of the ontological argument given by Anselm in Proslogion II is

“fallacious because it rests on the false doctrine that existence is a perfection,”

(Malcolm 1960, 44). However, he thinks that Kant, and all of his defenders in the 20th century have failed to show that the same is true for necessary existence. Similarly, objecting to the Cartesian version of the ontological argument does not, at least immediately, turn into an objection to Anselm’s argument, or any contemporary version of the argument. In fact, part of the reason for the resurgence of the ontological argument, aside from sociological conditions in academia, is the development in the early 20th century of various modal logics, and the possible world semantics often associated with these logics. The development of modal logic and semantics for it, according to people like Malcolm, Hartshorne, and Plantinga, allow us to rescue and reformulate some version of the ontological argument. So modal arguments are hermetically sealed off from certain kinds of criticisms because of their different structure and their usage of a system that allows us to be more precise about necessity, possibility, and the like. Note, though, that there are not only parody objections to definitional or conceptual ontological arguments, but also parody objections to any of

79 the modal arguments.19 Parody objections, then, are suited to just about any version of the ontological argument, and this breadth of applicability is one reason to evaluate them more highly than has been historically the case. In fact, this is one of considerations I argued for in section IV.2 regarding the supremacy of AntiGod parodies to other kinds of parody. There is certainly no hint of a straw man in

Nagasawa’ presentation; like proponents of parody arguments, he wants to take on the rival view on its strongest terms.

In order to discuss Nagasawa’s critique of Millican’s parody, we must examine both Millican’s formulation of Anselm’s argument, as well as his parody.

Millican’s version of Anselm’s argument is as follows.

1. The phrase ‘a-nature-than-which-no-greater-nature-can-be-thought’ is clearly understood by the Fool, and apparently makes sense. 2. Hence, we can take the phrase ‘a-nature-than-which-no-greater- nature-can-be-thought’ as successfully denoting some specific nature. 3. A nature which is instantiated in reality is greater than one which is not. 4. So, if a-nature-than-which-no-greater-nature-can-be-thought were not instantiated in reality, then it would be possible to think of a nature that is greater. 5. But this would be a contradiction, since it is obviously impossible to think of a nature that is greater than a-nature-than-which-no-greater- nature-can-be-thought. 6. Therefore, a-nature-than-which-no-greater-nature-can-be-thought must indeed be instantiated in reality. (Millican 2004, 457 – 458)

Millican’s parody argument contains the following two substitutions for

Anselm’s argument.

1a. The phrase ‘a-nature-than-which-no-more-effectively-evil-nature- can-be-thought’ is clearly understood by the Fool, and apparently makes sense. 2a. Hence, we can take the phrase ‘a-nature-than-which-no-more- effectively-evil-nature-can-be-thought’ as successfully denoting some specific nature. (Millican 2004, 461)

19 The language of definitional and conceptual ontological arguments is from Oppy 1995. 80

The parody moves straightforwardly from this point. Since it parallels the ontological argument, it is supposed to prove the existence of AntiGod just as well as the ontological argument proves the existence of God. But there is no such being as

AntiGod (for either theists or atheists). So the ontological argument must be fallacious.

Nagasawa contends that in order for this parody to be successful, Millican also needs to substitute the following premise for 3 in his formulation of Anselm’s argument. “3a. A nature which is instantiated in reality is more effectively evil than one which is not,” (Nagasawa 2010, 80). However, this is open to several immediate counterexamples. To use Nagasawa’s own, an existent morally perfect being would be more effectively evil than any nonexistent being. To make the substituted principle stronger, Nagasawa suggests it be altered to “3b. An evil nature which is instantiated in reality is more effectively evil than one which is not,” (Nagasawa 2010, 80). This change requires Millican to alter premises 4 and 5 of the original argument to the following, which allows us to get to the conclusion directly presented.

4a. So if a-nature-than-which-no-more-effectively-evil-nature-can-be- thought were not instantiated in reality, then it would be possible to think of a nature that is more effectively evil (for example, any evil nature that is in fact instantiated in reality). 5a. But this would be a contradiction, since it is obviously impossible for think of a nature that is more effectively evil than a-nature-than-which- no-more-effectively-evil-nature-can-be-thought. 6a. Therefore, a-nature-than-which-no-more-effectively-evil-nature-can- be-thought must indeed be instantiated in reality. (Nagasawa 2010, 80 – 81)

According to Nagasawa, these changes make the argument formally valid at the cost of losing both its structural and dialectical parallels with the ontological argument.

Regarding its structure, Nagasawa argues that the theory of natures that underlies the

81 original argument is concerned with greatness, not effective evilness. Regarding its dialectical efficacy, Nagasawa suggests both that no proponent of the ontological argument would find it reasonable to accept 3b, and that the argument is explicitly question-begging in ways the original argument is not. This is because Nagasawa thinks that the use of the term “effective” already includes an instantiation requirement, whereas greatness does not. So Millican’s argument fails to be either structurally or dialectically parallel to Anselm’s argument.20

Nagasawa also presents a version of the devil argument that he holds to be better than Millican’s AntiGod argument in regards to structure and dialectic.

1c. The phrase ‘a-nature-than-which-no-worse-nature-can-be-thought’ is clearly understood by the Fool, and apparently makes sense. 2c. Hence, we can take the phrase ‘a-nature-than-which-no-worse- nature-can-be-thought’ as successfully denoting some specific nature. 3c. A nature which is instantiated in reality is worse than one which is not. 4c. So if a-nature-than-which-no-worse-nature-can-be-thought were not instantiated in reality, then it would be possible to think of a nature that is worse (for example, any nature that is in fact instantiated in reality). 5c. But this would be a contradiction, since it is obviously impossible to think of a nature that is worse than a-nature-than-which-no-worse- nature-can-be-thought. 6c. Therefore, a-nature-than-which-no-worse-nature-can-be-thought must indeed be instantiated in reality. (Nagasawa 2010, 81 – 82)

This argument is “clearly structurally parallel: it merely replaces the phrase

‘greater’ in the original argument with ‘worse’,” (Nagasawa 2010, 82). Nagasawa thinks that even though this parody objection is closer to the structure of the ontological argument than Millican’s, it still fails in that it is not dialectically parallel.21 The devil

20 It is not clear to me why the argument being question-begging is a dialectical infelicity, as opposed to a structural one, but this hardly matters. 21 This is not how Nagasawa himself characterizes this problem. However, he says immediately before this that the argument is “clearly structurally parallel” and since the only other failure he mentions is characterized as failing to be dialectically parallel, this seems to be the appropriate label. 82 argument given above would not threaten a proponent of the ontological argument because 3c is inconsistent with 3. I am not sure why Nagasawa thinks that this removes the devil argument from consideration by proponents of the ontological argument.

While it is true that 3c is inconsistent with 3, the end result of either argument will be inconsistent with the conclusion of the other.

One way to understand what Nagasawa is saying here is to interpret him as charging the devil argument with an absence of dialectical parallel. The proponent of the ontological argument may simply feel no reason to accept 3c, given their previous commitment to 3. And perhaps this is correct. If I am antecedently convinced of A, and a premise in one of your counterarguments implies ~A, then of course I am likely to see you as simply begging the question against my argument. So, whether it is structurally infelicitous to have a premise that is inconsistent with the parodied argument, it may very well be dialectically inappropriate.

Finally, let us consider Nagasawa’s analysis of Gaunilo’s Island objection

(Nagasawa 2010, 88). He suggests that its rejection by proponents of the argument

(both historical and contemporary) is illustrative of two common problems with parody objections. For one thing, the Island fails because its scope is inappropriately narrow.

Instead of being concerned with the entirety of possible entities, the island objection is concerned only with possible islands. On Nagasawa’s view, this is evidence of

Gaunilo’s objection failing to be structurally parallel to the ontological argument. For another, the Island objection is committed to an assumption that proponents of the ontological argument are not committed to. While proponents of the argument are committed to there being some intrinsic maxima (i.e. upper limit or bound) about the

83 properties traditionally associated with God, they are not so committed to the idea that properties associated with islands also have an intrinsic maxima.22

Finally, we get a more determinate sense of what exactly the phrases

“structurally parallel” and “dialectically parallel” amount to. Nagasawa’s analysis of the various parodies discussed above always results in the parodies’ failure – either by failing to be structurally parallel, or by failing to be dialectically parallel.

These observations suggest the following hypothesis. The parody objection always fails because any parody argument is such that either (i) it is not structurally parallel to the ontological argument (typically because its scope is too narrow), or (ii) it is not dialectically parallel to the ontological argument (typically because it makes extra assumptions to which the proponents of the ontological argument are not committed). (Nagasawa 2010, 91)

While these remarks are helpful in determining the content of the phrases “structurally parallel” and “dialectically parallel” the next paragraph shows us just how far Nagasawa takes these notions!

To circumvent (i) and (ii), proponents of the parody objection need to revise either the no-devil argument or no-devil corollary, by eliminating the assumptions which proponents of the ontological argument do not endorse, while keeping the structure parallel to that of the ontological argument. However, ironically, once an instance of the parody argument is modified in this way, it is no longer a parody: it is the ontological argument itself. Of course one cannot undermine the ontological argument by advancing the ontological argument itself. (Nagasawa 2010, 91)

For Nagasawa, then, an argument being structurally parallel to the ontological argument means at least having the same scope as the ontological argument (all possible beings). There are other conditions, but these are very unclear. For example, Millican’s argument is charged with failing the test of being structurally parallel because it

22 As in note 7, this is not how Nagasawa himself characterizes this failure of the Island objection. But when we see how he discusses what “not being dialectically parallel” refers to, my interpretation will be confirmed. 84 replaces the property of “greatness” with “effective evilness,” (Nagasawa 2010, 80 –

81). But the devil argument given by Nagasawa is said to be “clearly structurally parallel” and the reason cited for this is that it “merely replaces the phrase ‘greater’ in the original argument with ‘worse’,” (Nagasawa 2010, 82). We will come back to this point.

An argument fails to be dialectically parallel when it either contains a commitment that is not held by the proponents of the paralleled argument, or when it contains circular reasoning not present in the paralleled argument (Nagasawa 2010, 81,

91).

According to Nagasawa, all parody objections necessarily fail, either by not being structurally parallel or by not being dialectically parallel. Finally, we should in fairness note that Nagasawa proposes this as a hypothesis, although his final statement evinces a higher level of certainty than hypothesizing would permit.

V.B Gaunilo’s Revenge

While I find parts of Nagasawa’s article persuasive, I think that both notions of

“structurally parallel” and “dialectically parallel” are problematically defined. First, I think that the notions of “structurally parallel” and “dialectically parallel” are too vague to be of any use here. In fact, I think that there are times that the terms act so elastically that one suspects some kind of circularity. Second, I think that the phrase “structurally parallel” contains an unreasonably high standard. It approaches identity, and this has to be false as two things can be structurally parallel without being identical.23 I will first

23 This seems obvious to me, but I will attempt to argue for it by carefully considering the notion of “parallel.” 85 examine and discuss both being structurally parallel and being dialectically parallel, and then give my arguments against the viability of these terms as used by Nagasawa.

Nowhere does Nagasawa provide us a clear and precise definition of either

“being structurally parallel” or “being dialectically parallel.” It is safe to assume that these terms are used in a rough but intuitive way. Regarding being structurally parallel, the structure of an argument might be thought to only be its syntactic properties.

However, this is too broad for obvious reasons of application. An argument A being structurally parallel to an argument B, at least in the case of parody arguments, is supposed to show something important about either the content or the processes used in

A or B. If the structural parallel only requires syntactic parallels, though, then arguments clearly unrelated to one another would be sufficiently parallel (and thus, possibly parodic). On this understanding, the following two arguments would be structurally parallel.

1d. If John gets a boo-boo, then John cries. 2d. John gets a boo-boo. 3d. Therefore, John cries.

And

1e. If apples turn orange in the winter, then Gore will be elected President of the Galaxy. 2e. Apples turn orange in the winter. 3e. Therefore, Gore will be elected President of the Galaxy.

I am certainly not suggesting that these two arguments do not share some structural elements (I doubt that Nagasawa would deny this either). They both exhibit the modus ponens style of argument. However, if the metric of being structurally parallel is to give us some information about whether the content of the argument, as

86 well as its form, is viable or not, or if it is to be of use in a parody style argument or objection, then it must be more strict. 1e-3e do not suffice as a parody of 1d-3d.

So far, Nagasawa and I are in supposed agreement. An argument being structurally parallel to another will require a high degree of similarity not only in its syntactic properties (how many premises is has, the logical structure of those premises, etc.) but also in its semantic properties. For example, recall that Nagasawa objects that

Millican’s argument is not “structurally parallel” to the ontological argument since it either a) 3 in the original argument is concerned with the property of “greatness,” whereas 3b is concerned with the property of “effective evilness,” or b) because the first part of 3 is about natures full stop, whereas the first part of 3b is about evil natures.24 In either case, the concern is not with broad logical structures of the argument, but the substitution or replacement of one term with another.

The problem here is that the line between two things having the property “being structurally parallel” and “not being structurally parallel” is never clearly defined, and when the statement is made that something is or is not structurally parallel with another thing, Nagasawa gives no defense of this.25 I find it highly implausible that a mere substitution of “worse” for “greater” throughout the argument counts as an instance of being structurally parallel, but the substitution of “effectively evil” for “greatness” throughout the argument is not (in fact, they seem to me to both count as structurally parallel). What could countenance such a judgment? If Nagasawa has a worked out

24 Disjunct b.) is discussed in Nagasawa 2010, 80, but Nagasawa says “This modification renders the parody argument not structurally parallel to Anselm’s ontological argument, but I set this point aside in favor of Millican.” Notice, though, that this does count as an instance of a failure of the parody argument to be structurally parallel to the ontological argument, so can aid us in attempting to understand the content of the phrase for Nagasawa. 25 I assume that the property “being structurally parallel” is a two place, relational property. So, for any argument x, any argument y, and the property “being structurally parallel” represented by Z, we would have Zxy, or “x stands in relationship Z to y.” 87 system for deciding these things, it is never shared with the reader, explicated, or defended.

Of course, this judgment looks even more suspicious when every instance of a structurally and dialectically parallel parody turns out to be harmless, while the potentially unsettling and effective objections are considered too dissimilar to be counted as structurally and dialectically parallel.26 Without being given more information, and without any additional argumentation on the part of Nagasawa, I think we must reject the usefulness of the property

“being structurally parallel” under Nagasawa’s description on grounds of vagueness

(and consequently, Nagasawa’s broad criticism of all parodies).

Regarding being dialectically parallel, we again are operating on a rough and intuitive level. To be charged with dialectically failing in some way, an argument or critique is made without taking into account its place in the dialectical structure. For example, I might have a greater burden to bear in justifying a premise of an original argument rather than one involved in a criticism, because if I am making the argument, then dialectically my position is such that it is important that I convince you or show my argument to be justified. The critic, on the other hand, is proving a negative rather than positive thesis, and so might be considered to be less constrained by the original commitments of the argument.

In the case of the parodist, they are trying to accept all or the majority of the controversial metaphysical assumptions involved in the argument. Of course, we wouldn’t expect the parodist to accept every assumption or consequence of the

26 I am getting a bit ahead of myself in speaking of arguments being dialectically parallel, but for the sake of accuracy in this sentence, I must. The reader will find this statement justified in time. 88 argument, especially if it ruled out the parodists’ own argument. We might judge an argument to be more or less dialectically parallel with another by determine exactly how many of the assumptions of the original argument are accepted by the proponent of the other argument.

Would we expect that the proponents of the original argument would also need to assent to every premise of the parody argument in order for it to be dialectically parallel? Nagasawa holds that this is the case, but it strikes me as plausible only in a stronger sense. Perhaps if we want the parody argument to be dialectically identical to the original argument, we would require this. But a things being parallel to another and a thing being identical to another are not, with respect, identical.

Of course, if we are conscientious parodists we will spend some time motivating those premises that bear a structural similarity to the original argument but that the proponents of the original argument do not accept. We might give long, extended arguments showing the similarity between the premises (save the necessary crucial difference that allows for a parodic conclusion) of the parody argument and the original argument. We might show how the traditional moves made in support of the premise in the original argument can also be made for the premise in our parody counterargument.

Or, of course, we might point out that a premise not argued for in the original argument does not commit us to make a long, extended argument for the similar premise in our parody argument. We might, as conscientious parodists, do any of these things. But, if at the end of the day, we found the proponents of the original argument resolute in their rejection of our similar but slightly altered premise, should this count as failing to be dialectically parallel? I do not think it should. Without additional explication of the

89 concept or argument from Nagasawa, I conclude we should reject his account of “being dialectically parallel to” as too vague, or (alternatively too strong).

An additional difficulty is captured when we reexamine the strength of

Nagasawa’s final characterization of how parody arguments could possibly overcome failure to either be structurally or dialectically parallel.

To circumvent (i) and (ii), proponents of the parody objection need to revise either the no-devil argument or no-devil corollary, by eliminating the assumptions which proponents of the ontological argument do not endorse, while keeping the structure parallel to that of the ontological argument. However, ironically, once an instance of the parody argument is modified in this way, it is no longer a parody: it is the ontological argument itself. Of course one cannot undermine the ontological argument by advancing the ontological argument itself. (Nagasawa 2010, 91, emphasis mine)

This characterization, coming at the end of the paper and after the preceding arguments, gives us the most penetrating insight into Nagasawa’s notions of “being structurally parallel” and “being dialectically parallel” yet. In fact, this restatement of how something would be both structurally and dialectically parallel sounds more like a definition of structural and dialectical identity. If, in order to be structurally and dialectically parallel to the ontological argument, parody arguments must essentially become the ontological argument, then the concepts we have been discussing are simply identity concepts, not concepts of being parallel to. If this is, true, though, then it begs the question against the parodist. If being structurally and dialectically parallel to the ontological argument comes to mean being structurally and dialectically identical to the ontological argument, then we should not be surprised that all parodies everywhere must always fail. We don’t even need to argue for this conclusion, rather we should just explicitly define the concepts early on in our exploration.

90

Just as it does not come as a surprise that all parody arguments must fail if the properties of “being structurally parallel to” and “being dialectically parallel to” are actually properties that require the two arguments to be identical and not merely parallel, it will come as no surprise that any parodist worth their salt will reject

Nagasawa’s account of these notions as not only vague, but also question begging. Of course, I will argue that the parody objections discussed above (especially the AntiGod objection) do, ultimately, fail, but not for the reasons that Nagasawa has given.

Section VI. Conclusion

In this chapter, I have argued that parody arguments are stronger than other kinds of objections to the ontological argument. I have modified and expanded Oppy’s taxonomy of parody objections to the ontological arguments, and explained the form of each. I have argued that an entire class of parody objections can be safely rejected by the proponent of the ontological argument. I have argued that, with respect to what are normally called “devil parodies” should instead be referred to as “AntiGod” parodies. I have argued that, of the remaining possible parodies to the ontological argument,

AntiGod parodies are the strongest and most effective available. Finally, I examine

Nagasawa’s argument that all parody objections necessarily fail. In doing so, I conclude that his global argument against parodies is poorly supported due to the vague and elastic nature of the notions of “being structurally parallel to” and “being dialectically parallel to”, since these two criteria are the main reasons given by

Nagasawa to reject all possible parodies.

If this chapter as a whole is correct, then the major obstacle to the ontological argument will be AntiGod parodies, parodies that purport to prove the existence or

91 necessary existence of a supremely powerful, supremely knowledgeable, and supremely evil immaterial being. Since this kind of objection is as powerful as I claim, a natural move towards both understanding its content and understanding whether it does serve as a true parody of the ontological argument will require engaging in some metaphysics of ethical concepts. In the next chapter, I examine the convertibility thesis of being and goodness (and the related privation theory of evil) held by medieval philosophers. If this account of the relationship between being and goodness, and goodness and evil, is correct, then we will be able to rule out parody arguments for AntiGod.

92

Chapter 4: Being and Goodness

Section I. Introduction

In this chapter, I will explore the convertibility thesis of being and goodness

(hereafter CT). At the beginning of the chapter, I will discuss some details of historical treatments of CT. Since it plays a significant role in the central argument of the dissertation (Chapter 5), it will be necessary to discuss CT at length, give possible formulations of the thesis, and give some content to the notions of both “being” and

“goodness,” all of which occur in section III. Of course, the most pressing concern is whether or not CT is viable for a contemporary analytic philosopher who is not antecedently convinced of the truth of theism. In other words, a critic might ask “Why should we think CT is itself true or even remotely plausible?" I will conclude by examining this question in some depth and arguing for the reflective plausibility of CT even in our own time. By “reflective plausibility” I mean that CT is plausible on reflection, after analysis and argument. I want to note at the outset that I am not attempting a full, robust defense of CT – if that were my aim, then that would be the entire focus of the dissertation. Instead, since so many contemporary philosophers believe that CT is so abjectly and obviously absurd as to require almost no consideration, I will aim only to demonstrate that CT has some basic level of plausibility and that the obvious considerations against it are neither obvious nor decisive. I will also show that there are arguments from Thomas Aquinas as well as contemporary writers in support of the thesis. This is what I mean by saying that I will defend the reflective plausibility of CT independently of theistic assumptions.

Section II. Being and Goodness – Historical Instances and Precursors

93

The origins of CT are located in a more general view of the relation between being and goodness. It has historically been argued that there must be some kind of necessary metaphysical relation between being and goodness, though in different works the relationships between particular characteristics are underspecified. Call this view proto-CT (hereafter PCT). PCT is found in the works of Plato and Aristotle, which influenced the views of later philosophers, especially those writing during the medieval period. This view is not as specific as CT, though CT is one particular instance of this more general view.

CT, simply put, is that being and goodness pick out all the same things in the world, even though the concepts are not identical. They are the same in extension but differ in intension. Put another way, they are the same in reference and differ in sense.

In the next section of this chapter I explain this thesis, but the primary focus of this initial section is to give the rudimentary historical context for this view.

The first instance of PCT stretches as far back as Plato.2728 For Plato, the world can be divided into the categories of material and immaterial entities (for example, the

Forms). Insofar as material entities change, degrade, come to be, and pass away, they are inferior to entities that are unalterable, incorruptible, and that always are, such as the

Forms. We already see that in Plato there are degrees of existence and that it is better to have the highest degree of existence rather than a lower degree. This is intuitively a

27 I'll note that since this dissertation is not focused primarily on engaging in historical scholarship, much of what I say here will be rough and general. There are undoubtedly issues of interpretation and language use that would either be too far afield for the topic of this chapter or that would require greater facility with ancient languages or scholarship than I possess. In my defense, I note that the view that idea that goodness and being have some meaningful connection in Plato and Aristotle is a common and mostly uncontested one in philosophy. For example, see Baker 2016, MacDonald 1991b, Reeve 2003, Schindler 2005, and Shields 2011. 28 All references to Plato are from Plato 1997. In text citations will follow the traditional citation style associated with Plato’s works of Title of Work, passage number and letter (e.g. Republic, 516a7-10). I will follow this same convention for Aristotle’s works, using Aristotle 1984. 94 statement of a relationship between being and goodness. However, in Plato, this goes even further. Asked how the forms are related to one another, Plato answers that the

Form of the Good is that which generates all other Forms, that which is the apex of existence or being (and by extension, that which generates all other objects) (Republic,

509b5-10). So at this point, we have a strong identification of the foundation of all being (and the highest being) with the Good. Although this is a different form of this relationship than that advanced by medieval philosophers, it is the basis for much of the later development of CT.

Aristotle, too, held a kind of PCT. Both goodness and being are

Transcendentals, according to Aristotle (Metaphysics IV, c. 2, 1003 b 23–4,

Nichomachean Ethics c. 4, 1096 a 12 sqq).29 This means that they are themselves not subject to the categories of substance, quality, quantity, etc. Rather, everything that can be classified by the categories exhibits the Transcendentals. One of the Transcendentals that is exhibited by everything that is, as given in the Metaphysics, is the good. Further,

Aristotle says that the good is spoken of in as many ways as being (Nichomachean

Ethics c. 6 1096 a 23-24). Insofar as everything that may be categorized is existent or is a being, then being or existence transcends the categories. The same is true of good.30

Building upon the works of Aristotle and Plato, CT is developed by several medieval philosophers.31 The earliest example is perhaps Augustine of Hippo, who argues for CT in the City of God, Confessions, and the Enchiridion. Insofar as existent

29 These particular passages are cited in Goris and Aertsen 2013. 30 For more on this connection, see Goris and Aertsen 2013. 31 Of course, the precise dating of the medieval period has been and continues to be a matter of controversy. For example, someone might object that Augustine is a Church father and not a medieval philosopher. The precise dating is of no importance to the author. Readers should alter the labeling as they see fit for Augustine, Boethius, and pseudo-Dionysius. 95 things flow from God, which is the First Good, they are good. Augustine also holds that for something to realize its good, it must first exist. The more fully it realizes its good, the more fully it exists. This is pointedly brought out when Augustine considers the nature of evil. Evil is classified as a privation, as opposed to something with its own positive character. For something to be purely evil, would simply be for that thing to not exist at all. He writes,

Therefore, if they shall be deprived of all good, they shall no longer be. So long, therefore, as they are, they are good; therefore whatsoever is, is good. That evil, then, which I sought whence it was, is not any substance; for were it a substance, it would be good. For either it would be an incorruptible substance, and so a chief good, or a corruptible substance, which unless it were good it could not be corrupted. I perceived, therefore, and it was made clear to me, that Thou made all things good, nor is there any substance at all that was not made by You; and because all that You have made are not equal, therefore all things are; because individually they are good, and altogether very good, because our God made all things very good. (Augustine 1887, Confessions VII.xii)

This leads us to a certain view of CT, though it is not the only view. In fact, it is not the only view of CT in Augustine’s own writings. For example, in City of God,

Augustine writes of the fallen angels

And to them it is an evil, solely because it corrupts the good of their nature. It is not nature, therefore, but vice, which is contrary to God. For that which is evil is contrary to the good. And who will deny that God is the supreme good? Vice, therefore, is contrary to God, as evil to good. Further, the nature it vitiates is a good, and therefore to this good also it is contrary. But while it is contrary to God only as evil to good, it is contrary to the nature it vitiates, both as evil and as hurtful. For to God no evils are hurtful; but only to natures mutable and corruptible, though, by the testimony of the vices themselves, originally good. For were they not good, vices could not hurt them. For how do they hurt them but by depriving them of integrity, beauty, welfare, virtue, and, in short, whatever natural good vice is wont to diminish or destroy? But if there be no good to take away, then no injury can be done, and consequently there can be no vice. For it is impossible that there should be a harmless vice. Whence we gather, that though vice cannot injure the unchangeable

96

good, it can injure nothing but good; because it does not exist where it does not injure. This, then, may be thus formulated: Vice cannot be in the highest good, and cannot be but in some good. Things solely good, therefore, can in some circumstances exist; things solely evil, never; for even those natures which are vitiated by an evil will, so far indeed as they are vitiated, are evil, but in so far as they are natures they are good. (Augustine 1887, City of God XII.iii)32

Thus, as MacDonald has also pointed out, Augustine seems to be inconsistent concerning his understanding and treatment of CT, as he either adopts different approaches at different times or confusedly merges the two together (MacDonald

1991b, 9-11). These different approaches will be discussed briefly at the beginning of

Section III in this chapter.

Boethius is another early proponent of CT. In fact, the development of later medieval thought is particularly influenced by Boethius’ work, as several commentaries on De Hebdomadibus (Boethius 1991) show. In Boethius 1991, we find that author pondering the question of how existent things are good, when they are not substantial goods. The answer, it turns out, is that they are good in virtue of the fact that they are created and sustained by the First Good, which is God.33 De Hebdomadibus is a short, dense, and above all opaque text. Consider the following passage:

Therefore, they are said to be good because their being flowed from the will of the good. For the first good, because it is, is good in virtue of the fact that it is. But a second good, because it flowed from that whose being itself is good, is itself good. But the being itself of all things flowed from that which is the first good and which is such that it is properly said to be good in virtue of the fact that it is. Therefore, their being itself is good, for it is then in it [--that is to say, the first good]. (Boethius 1991, 302)

32 There are many other accounts consistent with the varied view that Augustine has about CT and its corollary, the privation theory of evil. For example, Chapters III through VI of Augustine 1950 are about both CT and the privation theory of evil in different ways. 33 What MacDonald calls the “Participation Approach” to CT is captured in this account as well as some versions of CT in Augustine. It does make creatures rely on their being sustained by God in order for them to be good, and so this kind of property is relational instead of intrinsic. I thank Neal Judisch for this point. 97

It is difficult to see exactly how Boethius’ argument is supposed to go, but there can be no doubt that this work subsequently influenced a great deal of later medieval thought on being and goodness. Boethius’ understanding of CT is allied with some of the things that Augustine writes. Both Augustine and Boethius were heavily influenced by neo- platonic philosophers, and the emanationism in those philosophers’ approach undoubtedly shaped the views of Boethius and Augustine. A view of emanation where creatures and objects participate in God’s goodness and being, and where God sustains both their goodness and being, supports this version of CT.

The final early medieval philosopher that we will take note of in this section is pseudo-Dionysius, the Areopagite. Since pseudo-Dionysius, as Augustine and Boethius before him, was heavily influenced by neo-platonic thought, it is not surprising that his version of CT is also flavored by emanationism. In The Divine Names (Pseudo-

Dionysius 1987) pseudo-Dionysius asserts CT, or something like it, at several points.

As with Augustine, the content of CT for pseudo-Dionysius is brought out most powerfully when considering evil, its character, and its origin. In his work, we see that the Good is a precondition for existence of any kind. Good, then, will always be co- extensive with being, and vice versa as the cases of supposed evil things illustrate. He writes,

Evil does not come from the Good. If it were to come from there, it would not be evil. Fire cannot cool us, and likewise the Good cannot produce what is not good. If everything comes from the Good—and the Good naturally gives being and maintains, just as evil naturally tries to corrupt and destroy—then no being comes from evil. Nor will evil itself exist if its acts as evil upon itself, and unless it does this, then evil is not entirely evil but has something of the Good within it which enables it to exist at all. (Pseudo-Dionysius 1987b, 84 – 85)

98

Finally, as a historical note, I should point out that many other philosophers held this view, especially leading up to and continuing after Thomas Aquinas, who will be a focal point for the next section of this chapter. However, the purpose of discussing each of these figures is not to give an extensive, detailed account of the history of CT or its forebears. Rather, it is to note some important, influential accounts of CT or PCT that influenced Aquinas and later writers, and to give the reader an idea of exactly how far back in history this tradition stretches.

Section III. Being and Goodness – An Exploration of CT in the Thought of St.

Thomas Aquinas

In this section, I describe Aquinas’ influential account of CT in detail.34 In the first place, though, we need to examine the two different versions of CT and focus our attention on the version that Aquinas adopts. This is especially important for the following purposes of the chapter and the remainder of the dissertation since the Nature

Approach (most thoroughly developed by Aquinas) is the only viable approach to adopt in the dialectical context of the ontological argument. This is because adopting the

Participation Approach begs the question and explicitly assumes the existence of God, which undercuts its effectiveness as a background conceptual scheme or theory to support an argument for the existence of God.35

III.A The Participation and Nature Approach to CT

34 Aquinas develops this most carefully in ST I, Q 5, A 1. I rely heavily on secondary sources for my understanding of Aquinas’ metaphysics of being and goodness, especially Aertsen 1985, 1991, 1996, 1999, Elders 1993, Geach 1955, Goris and Aertsen 2013, Kenny 2002, McInerny 1991, Stump and Kretzmann 1991, several essays from MacDonald and Stump 1999, as well as the rest of MacDonald 1991 not already included in this list. 35 It should be noted that Aquinas rejects the efficacy of Anselm’s ontological argument in ST I, Q 2, A 1, ad. 2. This is not related to his version of CT, but for epistemic and conceptual reasons discussed in the passage cited above. 99

MacDonald argues that the place of CT in medieval thought is complicated by the fact that there are two different, yet prominent ways of understanding the relation between being and goodness. On the Participation Approach, favored by Boethius, pseudo-

Dionysius, and (at some points) Augustine, good and being are co-extensive because of the relationship between God and created things. God is the first, foundational good.

He is the source of all things in existence, and consequently the source of all things that are good. Since God is the ultimate good, and evil cannot come from good any more than cold from heat, then the things God creates must be themselves good. This approach draws on a principle by which a created thing relies on its creator for its qualities. As God is good, and all things come from God, then because they are created and sustained by the ultimate goodness, they are themselves good. This approach was popular among Neo-Platonist philosophers and accordingly is considered to be a development of the thought of Plato.

Another approach, found in (at times) Augustine, and most powerfully in

Aquinas, is what MacDonald calls the Nature Approach. On this view, being and goodness are co-extensive because of the relation between a thing and its completed or perfected state. Insofar as being is a prerequisite for the fulfillment of a thing’s nature, it is good. Alternately, insofar as all good things are, they have being.

The Nature Approach can be traced to Aristotle. For Aristotle, a thing is a good

X, in proportion to how far it attains the perfected or completed nature relevant to the kind of thing it is. For example, we call a hammer good in proportion to how well it fulfills its specifying function, in this case hammering nails. We would call a hammer bad when it is broken, poorly made, or otherwise deficient in attaining the end

100 distinctive to it, again hammering nails. Although the story gets more complicated as we move from artifacts to natural creatures, the same thesis holds. Something is good just as far as it fulfills the specifying capabilities unique to its kind.

The preceding discussion of the Nature Approach is rough; however, a detailed examination of Aquinas’ and Alexander’s accounts of CT will clarify the approach considerably.

III.B Aquinas and CT

Aquinas’ version of CT is one of the clearer examples of the doctrine found among medieval philosophers. However, to discuss Aquinas’ account of being and goodness is to wade into the massive, sprawling territory of his metaphysics. I will not attempt a general account of Aquinas’ metaphysics here, both for reasons of brevity and competence.

First, it will behoove us to discuss some of Aquinas’ arguments for the convertibility of being and goodness.36 The first argument occurs in the subsection of the Summa Theologiae, Part I, Question 5, Article 1 entitled “Are goodness and being the same really?” (Aquinas 1920, ST I, Q 5, A 1). We will refer to it as the Argument from Desire. Here, Aquinas begins by asking whether goodness differs really from being. He considers this question by way of three initial objections: that to perceive goodness and to perceive being are separable, that good has the form of being and that nothing can be its own form, and that goodness come in degrees, unlike being. I will give Aquinas’ answer to this question in full, and then move through it step by step.

36 A good, though short, discussion of these arguments can be found in Stump and Kretzmann 1991, pp. 99 –101. I am indebted to this work especially in aiding my understanding of Aquinas’ metaethics, as well as his arguments for the convertibility thesis. 101

Goodness and being are really the same, and differ only in idea; which is clear from the following argument. The essence of goodness consists in this, that it is in some way desirable. Hence the Philosopher says (Ethic. i): "Goodness is what all desire." Now it is clear that a thing is desirable only in so far as it is perfect; for all desire their own perfection. But everything is perfect so far as it is actual. Therefore it is clear that a thing is perfect so far as it exists; for it is existence that makes all things actual, as is clear from the foregoing (ST I, Q 3, A 4; ST I, Q 4, A 1). Hence it is clear that goodness and being are the same really. But goodness presents the aspect of desirableness, which being does not present. (Aquinas 1920, ST I, Q 5, A 1, Co.)

The first premise of the argument repeats a familiar Aristotelian theme. It is that

“goodness is desirable.” To put it better, we might say: “Whatever is good is desirable.” Aquinas then states that "Something is desirable only insofar as it is perfect." We can already see where the argument is going, and Aquinas immediately answers the question bound to be on our minds. He states that "Everything is perfect so far as it is actual." Then, we get Aquinas' conclusion. "Therefore it is clear that a thing is perfect so far as it exists… Hence it is clear that goodness and being are the same really. But goodness presents the aspect of desireableness, which being does not present.” More formally, we can state the initial version of the argument as follows.

1. The good is that which all desire. 2. Something is desirable only to the degree that it is perfect. 3. Something is perfect only to the degree that it is actualized.37 4. Therefore, a thing’s goodness is the same (in reality/in reference) as its being.

Aquinas quickly notes that, although goodness and being are really the same (or perhaps better, in reality the same) they do differ in conceptual content. He notes, immediately after establishing that being and goodness are the same, that the good

37 In premise 2 and 3, I take it that "insofar as" is synonymous with "to the degree that" and "to the extent that." I think that this is a natural reading of the phrase, but since it does alter the language used by Aquinas himself, it should be noted. 102 presents itself as desirable, which is not true of being. Thus, they cannot be conceptualized as identical, even if they are the same really.38

At first blush, the argument is too quick. In replying to the various objections, though, Aquinas aids our understanding of the argument put together in such sparse and straightforward terms. For example, in the Reply to Objection 1, Aquinas immediately makes a significant distinction that is not made explicit in the main text of the argument.

Substantial being, Aquinas says, is what a thing has once it is actual, as opposed to potential. This is similar in many ways to contemporary accounts of existence, where a thing either exists or does not. Insofar as something is in act, as opposed to in potentia, it has being simply. As another way to put this, we might say that if something is, it has being simply. This contrasts with what Aquinas calls a thing's relative being, which it has only as far as it is some thing or another. Aquinas uses the example of being white.

To be white is to be in some particular way or mode. To be white does not bring something out of potential, but rather something must already have being simply to participate in this more particular or precise mode of being. Following Jan Aertsen, we will call being simply B1, and being relatively B2 (Aertsen 1999).

Aquinas makes a similar distinction between good simply and good relatively.

To be good simply is to have achieved the ultimate perfection that is suitable for a thing, determined by its kind. On the other hand, to be good relatively is characterized by being only partially along the way to achieving the perfection suited to the thing in question. Aquinas writes “Hence that which has ultimate perfection is said to be simply good; but that which has not the ultimate perfection it ought to have (although, in so far

38 This way of wording the relation expresses the convertibility of being and goodness as opposed to the identity of being and goodness. 103 as it is at all actual, it has some perfection), is not said to be perfect simply nor good simply, but only relatively,” (Aquinas 1920, ST I, Q 5, A 1, ad. 1.). Here we can see that there are different degrees of goodness, and that the apex of goodness for a thing is actually labeled differently than its goodness in any degree short of perfection. Further, we may assert (following Aquinas) that simple existence is itself a relative good. How can we assert this? For Aquinas, a thing must exist simply before it can come to attain its full perfection, or its ultimate or simple goodness. Since a thing’s existence is the first step to attaining its perfection, its existence is itself a perfection (though the perfection of simple existence does not entail a thing’s ultimate perfection or goodness).

Again following Aertsen, I will call simple goodness G1, and relative goodness G2

(Aertsen 1999).

Using the distinctions between B1, B2, G1, and G2, we can make Aquinas’ CT in this section of the Summa Theologiae more precise. In fact, such a precisification follows what Aquinas says in the rest of the Reply to Objection 1. Insofar as something has B1, it also possesses G2, since G2 is a prerequisite for G1 and is entailed by B1.

Alternately, insofar as a thing has G1, it will have B2, since for a thing to reach its ultimate perfection will require it to be in some specified mode or another (such as the specified mode of being human). So, whether from examining something’s being simply, or something’s simple goodness, we arrive at a form of CT. “To be (B1) is to be good (G2),” and “To be good (G1) is to be (B2).” This specification of the differences between simple being and goodness, and their relative counterparts is helpful in further fleshing out why being and goodness, though convertible in reality, are nevertheless not convertible in conceptual content.

104

The second argument that Aquinas gives for CT can be found in both Summa

Contra Gentiles, Book I, Chapter 37 (Aquinas 1955, SCG, Bk. I, ch. 37), and Summa

Theologiae, First Part of the Second Part, Question 94 (Aquinas 1920, ST I-II, Q 94).

We will refer to it as the Argument from Preservation. Aquinas writes

Furthermore, “the good is that which all things desire.” The Philosopher introduces this remark as a “felicitous saying” in Ethics I. But all things, each according to its mode, desire to be in act; this is clear from the fact that each thing according to its nature resists corruption. To be in act, therefore, constitutes the nature of the good. Hence it is that evil, which is opposed to the good, follows when potency is deprived of act, as is clear from the Philosopher in Metaphysics IX. (Aquinas 1955, SCG, Bk. I, ch. 37)

And again, Aquinas writes that “Because in man there is first of all an inclination to good in accordance with the nature which he has in common with all substances: inasmuch as every substance seeks the preservation of its own being, according to its nature,” (Aquinas 1920, ST I-II, Q 94, A 2, co.). In both of these passages, we can make out an interesting connection between being and goodness.

1. The good is that which all things desire. 2. Each thing, according to its nature, resists corruption. 3. Each thing, according to its nature, desires to be completely actualized. 4. To be completely actualized is to be good. 5. Therefore, a thing’s goodness is the same (in reality/in reference) as its being.

This argument, as well as the Argument from Desire, begins with the

Aristotelian and Platonic conception of the good. Next, Aquinas states that all things desire to be completely actualized, as we can see from all things resisting their own corruption, or attempting to preserve their being. But if the good is to be identified with that which all things desire, and all things desire the complete or entire actualization of their nature, then the good must be identified with the complete or entire actualization

105 of a thing’s nature. Therefore, the good is the same in reality or in reference, as being.

This argument is more straightforward than the Argument from Desire, and I note that

Aquinas does not, in giving this argument, make the distinctions between B1, B2, G1, and G2 as above. However, I think that the argument could be modified in this way with no loss of meaning. In particular, I think that keeping these distinctions in mind is helpful for understanding Aquinas’ development of CT, even if he does not explicitly use this terminology in this argument.

The third argument in Aquinas’ vast corpus is to be found in both Summa

Contra Gentiles, Book III, Chapter 3 (Aquinas 1955, SCG, Bk. III, ch. 3), and in

Summa Theologiae, First Part of the Second Part, Question 1 (Aquinas 1920, ST I-II, Q

1). These sections are longer than the ones previously given, so I will not reproduce them here. Rather, I will give what I take the argument to be as it spans both works. I also note that this argument has significant similarities to the Argument from Desire.

We will call it the Argument from Ends.

1. The good is the final end of desire. 2. Ends are desired for themselves, or for their instrumental value in reaching a more significant end, or for both. 3. Whatever end is desired only for its own sake is whatever the desirer perceives to be its final end. 4. But, whatever is desired only for its own sake is the complete and entire fulfillment of a thing’s nature. 5. The complete and entire fulfillment of a thing’s nature is just its becoming fully actualized. 6. To be fully actualized is desired for its own sake. 7. The good is desired for its own sake. 8. Therefore, the good is the same (in reality/in reference) as being.

Again, the distinctions between B1, B2, G1, and G2 are not made explicit here, but we can alter the argument to include these terms with no loss of meaning. As I mentioned in discussing the Argument from Preservation, I believe that these divisions between

106 different ways of talking about being and goodness are helpful in understanding

Aquinas' view as a whole.

The final argument we discuss here is misleadingly short, compared to the others. It can also be found in the Summa Contra Gentiles Book III, Chapter 3

(Aquinas 1955, SCG Bk. III, ch. 3). We can call this the Argument from Being. He writes,

Besides, every action and movement are seen to be ordered in some way toward being, either that it may be preserved in the species or in the individual, or that it may be newly acquired. Now, the very fact of being is a good, and so all things desire to be. Therefore, every action and movement are for the sake of a good. (Aquinas 1955, SCG Bk. III, ch. 3.4)

We can put this in premise-conclusion form as follows.

1. Every action and movement are ordered toward being. 2. To be ordered toward being is to preserve the being of the species of an individual, or that being may be newly acquired. 3. To be ordered toward X is to desire X. 4. So all things desire being. 5. The good is that which all things desire. 6. Therefore, the good is the same (in reality/in reference) as being.

What do these various arguments demonstrate? First, the arguments demonstrate that CT is of central importance for Aquinas. It plays a significant role in

Aquinas’ system, and so spending time developing and arguing for this theory is important insofar as it will have widespread ripple effects on other parts of Thomistic philosophy. Second, that instead of merely adopting or positing the view, Aquinas gives several arguments for the Nature Approach to CT. Other ancient and medieval philosophers do argue for their version of PCT and CT, but Aquinas' style makes his arguments much clearer for present-day readers.39

39 Compare any of the arguments given by Aquinas and De Hebdomadibus, for example. 107

In this section, I have shown that Aquinas thoroughly develops the Nature

Approach of CT. I have also given several arguments for this approach as drawn from

Aquinas. The next section goes further by examining and explaining CT in a way that is not drawn directly from Aquinas, but from my own thoughts on the subject as well as how we might understand this view as applied to contemporary analytic philosophy.

Section IV. CT, Implications, and Formulations

In this section, I examine the Nature Approach to CT, as drawing from the framework developed by Aquinas. Then, I briefly discuss the distinction between metaphysical and moral goodness. As metaphysical goodness is a broader category than moral goodness, I will show that in some cases metaphysical goodness includes moral goodness, while in other cases it does not. Finally, I discuss the idea of degrees of being, and of unqualified and qualified being.

Since CT is central to the argument of the final chapter, and since it is notoriously difficult to explain, this section represents my own attempt to explain the nuances of CT and some implications.40 I have explained CT at several points in earlier sections, and I will assume the reader understands how being and goodness are supposed to relate on CT. In preparation for work completed in later sections, I note that qualified and unqualified being and goodness are captured in the G1, G2, B1, B2 schema used above in the discussion of Aquinas’ arguments for the Nature Approach of

CT, and taken from Aertsen 1999.

IV.A Metaphysical vs. Moral Goodness

40 In writing this section, I am deeply indebted to Alexander 2012, and those scholars that aided in my understanding of Aquinas’ Nature Approach to CT referred to in footnote 5. I have not attempted to make everything I say here consistent with Aquinas’ particular religious version of CT, but I do believe it to be in the spirit of a broadly Aristotelian/Thomistic Nature Approach to CT. 108

There are notions of goodness that seem to be entirely different from the familiar notion of moral goodness. In this section, I will briefly explain how I understand metaphysical goodness as it differs from moral goodness.

Moral goodness is a familiar concept to all philosophers. When we observe a person hurting another, or aiding another, we evaluate these situations in many ways: we might talk about praise, blame, intentions, consequences, character, obligations, rights, etc. Moral goodness is a familiar concept with a variety of applications.41 On the Nature Approach of CT, this will be analyzed in terms of the fulfillment or impediment of any entity’s nature or function, as long as that entity is a proper subject of moral consideration.

Metaphysical goodness, on the other hand, has a strange ring to it. To meaningfully talk about the goodness of nonhuman animals or artificial objects concerning their nature or function, we cannot simply use the sense of the phrase "moral goodness." Whether a predator is good at predating, or a thermometer good at measuring temperature is not a matter to which moral goodness or its opposite even applies.

By “metaphysical goodness” I mean any kind of goodness that an object or creature exhibits in virtue of its fulfilling its function or nature. A sub-species of this goodness will be moral goodness, for human beings (and other suitable objects of moral consideration). As a human being, if I am able to fulfill my function that will consist in reaching a certain state of moral as well as non-moral goodness. The remaining kinds of goodness that do not fall under the term “moral goodness” will be called “non-moral

41 Of course, this does not mean it is simple or settled – there are a number of interesting debates regarding the source and nature of moral goodness, the objects to which it can be appropriately applied, etc. But the concept is familiar enough for our purposes as to require no further time belaboring the issue. 109 goodness.” On the view sketched above, non-moral goodness can come in two varieties, natural and artificial non-moral goodness.

For natural non-moral goodness, consider the example of a good oak tree. The things that lead to evaluating an oak tree as a good oak tree are not artificially designed by human beings. Features such as speed of growth, efficiency in drawing nutrients from nearby sources, fecundity in dropping acorns, all of these things allow an oak tree to fulfill its function to a greater degree. And yet, it seems bizarre to talk about an oak tree’s being morally good in virtue of its having an efficient root system or being particularly fecund. This would count as metaphysical goodness as I’ve described it since each of these features allow it to fulfill its function, but would not to count as moral goodness.

For artificial non-moral goodness, consider the example of a good computer. A good computer is one that runs efficiently, has appropriate storage capacity, is free from viruses or malware, and has appropriate programs for our purposes, whatever those might be. We could further specify what makes for a ‘good computer,' depending upon the function it is designed for (in this case, meeting some specification of human needs and/or wishes). But none of the traits required for the computer to fulfill its artificially determined function will count as morally good ones. As before, to the extent that the computer has a function, even if designed by humans, it will be evaluated as more or less metaphysically good in virtue of those features that allow it to fulfill its function, but this will not count as moral goodness.

The point here is simply that on the Nature Approach to CT, we can meaningfully talk about the goodness of all objects with natures, including analyzing

110 non-moral natural and artificial goodness. This is a strength of the view – it allows for a unified treatment of these varying kinds of goodness. Next, let us turn to explaining degrees of goodness on this view.42

IV.B Degrees of Goodness, Qualified vs. Unqualified Goodness

Now, let us turn to an implication of CT. It is the notion that both being and goodness can come in degrees. For goodness, moral and non-moral, this view needs no defense, so I will spend little time on it here.

Different human beings have different degrees of moral goodness – most of us are somewhere in between Pol Pot and Gandhi. Similarly, we all know people who have some moral quality to a greater degree than we do and people who have some moral quality to a lesser degree than we do.

Degrees of non-moral goodness are familiar as well, both in humans and in non- human natural objects and artificial objects. When a person upgrades their computer after many years, it becomes very obvious very quickly that the new computer has a greater degree of goodness related to fulfilling computer functions. Similarly, insofar as the oak tree from our previous example is efficient at extracting nutrients (but not as much as it could be) or good at producing acorns (but not as much as it could be), it will have some degree of non-moral natural goodness. All of the preceding is straightforward, and I will spend no further time on it. However, to flesh out the Nature

42 This is an extremely rudimentary sketch of the metaethics of the Nature Approach to CT. There is a whole range of sophisticated metaethical literature on varieties of goodness, moral vs. natural goodness, the relationship between them, etc. It would be an interesting and worthwhile project to further develop the Nature Approach of CT as a robust, contemporary metaethical theory that would take into account contemporary issues, distinctions, and dilemmas in metaethics. However, I cannot pursue that project any further here, both for reasons of space and purpose. 111

Approach to CT, we will examine a distinctly Thomistic notion – qualified and unqualified goodness.

While this distinction borrows from the account of CT in the work of Thomas

Aquinas, one need not be a Thomist to accept this analysis. One simply has to find the distinction a plausible one. What does it mean for something to be unqualifiedly good, versus qualifiedly good?

Using Aquinas’ discussion as a starting point, we can call something qualifiedly good insofar as it has some degree (but not the highest degree) of the good expected or required of it. Recall that the metaphysical good of something is directly proportional to the degree it fulfills its function, or expresses its nature. However, every actual thing expresses its nature to some degree, by virtue of being existent (the first step, we might say, on the road to the full expression of one's nature). So every existent thing is good to some degree. Therefore, being qualifiedly good is just to have some degree of the goodness associated with the nature of a thing. Unqualified goodness is, as might be expected, the fully exemplified goodness of something that fulfills its function to the utmost degree possible for it. A human being at the height of fulfilling its function, or at the apex of exemplifying its nature, then we can say that human being is an unqualifiedly good human being, as we use the term here.

As I noted earlier, in some sense even unqualified goodness is qualified. What do I mean? Well, I take it that the goodness for human beings consists in a different kind of goodness than the goodness for cats, chairs, bugs, or spark plugs. The unqualified goodness of a chair will require properties such as sturdiness, comfortability, appropriate height or width for seating normal human beings, and

112 durability. The unqualified goodness of human beings will consist of different properties such as bodily health, mental health, rationality, and virtuousness. In this sense, then, the unqualified goodness associated with a human being is not identical with the unqualified goodness associated with a chair. And so on, and so on, for all sorts of other artificial and natural objects. This is what I mean when I say that even unqualified goodness is in some sense qualified, as in each case it will be relative to (or qualified by) the kind of thing to which it relates. That does not mean that the distinction is unhelpful, but rather that we must have a clear understanding of the concepts in order to avoid confusion.

IV.C Degrees of Being, Qualified vs. Unqualified Being

Now we turn to the more difficult corollary—showing that being also admits of degrees. On the face of it, this looks preposterous. Existence, we are told by followers of Frege and Russell, is a binary and univocal predicate. Something either exists, or it does not. What might it mean to talk about degrees of existence? How can we say, for example, that some existent thing x has a greater degree of existence than some other existent thing y? Part of this results from treating existence as Frege, Russell, and their followers would, but we have seen in the second chapter that this way of understanding existence, as both binary and as univocal, does not exhaust the logical space. There is, however, great intuitive appeal to the idea that there are not degrees of being, but that something is, or is not.

There are all sorts of things that we can safely say do not exist, in the traditional sense—winged horses, a golden mountain, 200-foot tall nuclear-fire-breathing saurians, satyrs, square circles, Superman, etc. Some of these entities are mythological, some

113 fictional, some of them contingently fail to exist, while others necessarily fail to exist.

On the other hand, there are existent entities—the phone on my desk, Barbara Walters, my 2000 Dodge Dakota truck, Robert Audi, my roommate’s dog Zoey, etc. These entities all contingently exist, but there might be necessarily existent entities (if the ontological argument is sound, or if traditional Platonism is true, or if Platonism concerning numbers is true, etc.). Where are the partially existing entities? Would these pop in and out of existence in accord with some set of metaphysical laws? Would they have a shadowy, slightly translucent appearance? What would it mean for something to have some degree of existence in between existence and non-existence?43

This question can be answered by proposing a conceptual scheme in which degrees of being or existence makes sense, and this scheme will likely involve a departure from our everyday, binary notion of existence. To unpack this substantially different way of speaking about being or existence, we will examine qualified and unqualified being. Consider the following way of treating the concept of being or existence.

Following Aquinas, we may distinguish qualified and unqualified being.

Unqualified being, or what some scholars call “being absolutely” or “being simpliciter” is something akin to the binary notion of existence. Every existent entity has unqualified being. Myself, this table, this computer, the curry I will soon eat, all unqualifiedly are, insofar as they exist, or are actual. The notion of qualified being, or relative being, is being in addition to the way in which I am, or the way in which a thing is. My qualified being includes such properties as being human, being male, being

43 While we have explored this distinction as it occurred in Aquinas, I develop it further here to demonstrate its usefulness as a contemporary account. 114 white, etc. The most important qualification of my being is the primary kind of thing I am, which is human. Insofar as I am more or less human, I exemplify my relative being to a greater or lesser degree. Whatever we think is unique to the nature of human beings (what we might call human nature) is not evinced equally in all humans. Insofar as I fail to be rational in any area of my life, I show a lesser degree of humanness. But when I fail to instantiate the unique aspect of humanity, I have a lesser degree of relative being. If virtuous conduct is unique to human beings, then at any point that I fail to be virtuous, I exhibit a lesser degree of relative being, being human in this case.

An existent tiger exhibits unqualified being, insofar as it is. On the other hand, it exhibits greater or lesser degrees of qualified or relative being just in case it fulfills its nature. Since a tiger is by nature a predator, we can talk about different tigers having greater or lesser degrees of being depending upon how effectively they conform to their principal nature. A tiger that is slow, cannot climb trees, has broken claws or teeth, will evince a lesser degree of its qualified being than a tiger that is fast, can climb trees, and has sharp claws and teeth. And so on for other animals, or plants, and (if this view is correct) everything in existence.

Although we have only considered two examples, of human beings and nonhuman animals, we can show that this treatment can be extended to all other things.

My chair, for example, has both a qualified and an unqualified sense of being attached to it. Unqualifiedly, it is, for if it were not I would fall to the floor (more likely I wouldn’t be seated in the first place). But it also has the qualified being of being a chair, of having a nature (even if created by us) to fulfill. When a chair supports me, when it is comfortable, when it allows an appropriate seated posture, it exhibits a

115 greater degree of qualified being. If a chair fails to support me, or is uncomfortable, or is such that my knees touch my elbows, it exhibits a lesser degree of its qualified being.

Now, let us turn to a contemporary account of goodness that allows for both metaphysical and moral goodness, and an important and controversial corollary of

CT—the privation theory of evil (hereafter PE).

IV.D A Contemporary Nature Account: The Attributive Account of Goodness44

An account of goodness that meshes nicely with the previous distinction of metaphysical and moral goodness, and that is compatible with CT is the attributive account of goodness. Alexander, in Goodness, God, and Evil (Alexander 2012) defends and develops a claim first made by Peter Geach, that “good” and “bad” are logically attributive adjectives rather than logically predicative adjectives (Geach 1956). Since this is a recently developed account that is consistent with CT in the form of the Nature

Approach, and since this account also implies PE, a way to understand evil that is also implied by CT, it will enrich our understanding of CT.

According to Geach, there is a distinction to be made between logically predicative and logically attributive adjectives (Geach 1956, 33). Logically predicative adjectives pass what Alexander calls the Splitting Test (Alexander 2012, 32). For example, “green” is a logically predicative adjective. When I say “x is a green turtle”

44 It is obvious why one would find the Nature Approach to CT more plausible, both as a metaethical account and as a background conceptual scheme that makes way for the ontological argument, as I have previously argued. Any Participation Approach immediately begs the question with respect to the ontological argument or any argument for God’s existence, and this is because at base the view explains the relationship between being and goodness by positing God. Further, many philosophers and biologists talk about human beings as well as other forms of life having natures, without expressly positing the existence of God. This suggests that the nature approach will be stronger as both a metaethical theory (since its central category, that of a nature, is a familiar and useful concept in contemporary biology and philosophy), and as a theory of great-making properties used in the ontological argument (since it does not beg the question before the argument gets off the ground). For examples and explanations of accounts of humans and other things having natures, none of which presuppose the existence of a God, see Buss 2001, Foot 2001, Hursthouse 1999, Nussbaum 1992 and 2000, and Stenmark 2009, 2012a, and 2012b. 116 we can infer from the truth of this statement both of the statements “x is green” and “x is a turtle.” Similarly, when I say “y is a rectangular quilt” we can infer from the truth of this statement both that “y is rectangular” and “y is a quilt.” Logically predicative adjectives also pass what Alexander calls the Higher-Order, Lower-Order Kind Test

(Alexander 2012, 33). If something is in a class that is itself part of a higher-order class

(or a lower-order class), then a predicative adjective will hold true for both of the classes. For example, from “x is a green turtle” and “all turtles are animals” we can infer that “x is a green animal.” This works in the opposite way too. From “y is a rectangular blanket” and “y is a quilt” we can infer that “y is a rectangular quilt.”

On the other hand, logically attributive adjectives fail both of the previous tests.

“Small” is a good example of a logically attributive adjective. From “x is a small bear” we cannot infer both “x is small” and “x is a bear.” Even a very small bear is quite large as compared to many other things of this world. Another example would be

“fast.” From “y is a fast sloth” we cannot infer both that “y is fast” and that “y is a sloth.” In both of these cases, the logically attributive adjectives fail the Splitting Test.

Logically attributive adjectives also fail the Higher-Order, Lower-Order Kind Test.

Using the previous examples, from “x is a small bear” and “all bears are animals” we cannot infer that “x is a small animal.” The point is while logically predicative adjectives always pass this test, logically attributive adjectives do not necessarily pass this test.

According to Geach and Alexander, “good” and “bad” are both logically attributive adjectives. From “x is a good writer” we cannot infer both that “x is good” and that “x is a writer.” Similarly, from “x is a bad bowler” we cannot infer both that

117

“x is bad” and that “x is a bowler.” In addition to failing the Splitting Test, “good” and

“bad” also fail the Higher-Order, Lower-Order Kinds Test. From “x is a good writer” and “all writers are humans” we cannot infer that “x is a good human.” Again, from “x is a bad bowler” and “all bowlers are human” we cannot infer that “x is a bad human.”

The point of both Alexander and Geach’s treatment of the attributive adjective

“good” is, at least in part, that for something to be good is for it to be good in reference to whatever kind of thing it is, and that this in turn suggests we need to know its function to make such a judgment. Geach writes

… if I do not know what hygrometers are for, I do not really know what 'hygrometer' means, and therefore do not really know what 'good hygrometer' means; I merely know that I could find out its meaning by finding out what hygrometers were for. (Geach 1956, 38)

There is, therefore, a connection between a thing’s goodness and its nature. As stated before, the fact that this account of goodness is connected with the idea that things have natures is advantageous for our purposes. Alexander argues that the phrase

"x is good" is always either incomplete or elliptical for “x is a good being of kind K,” following Geach (Alexander 2012, 35). He also provides an analysis of what it means for something to be a good being of a certain kind, and this analysis is compatible with the foregoing account of the Nature Approach to CT.

In fact, the relation between the attributive account and CT is stronger than compatibility. Alexander argues that the attributive account actually implies the truth of

CT in two ways. The first way only seems to imply the truth of a certain form of CT, the Nature Approach. The second way focuses on PE, which Alexander argues implies the truth of CT (this is interesting, as most often the implication relation between CT

118 and PE goes the other way around). Since our next section is focused on PE, we will consider this argument only briefly before exploring that most controversial part of CT.

The first argument focuses on the way the attributive account links goodness with function. There are two parts to showing the connection between the attributive account of goodness and CT. On Alexander’s account, the goodness of a thing necessarily depends upon the nature of that thing, and what the function of that thing is.

To be a good X is to have certain features that are characteristic of the kind of thing that

X is, to some specified degree. What counts as a success or failure in being a good member of a certain kind will depend intimately on the nature of the kind of thing under consideration. Insofar as something more fully fulfills its function or more fully exemplifies its nature, it is a better member of its kind than one that fails to fulfill its function or exemplify its nature. However, a thing fulfilling its function or exemplifying its nature is simply a matter of it having the appropriate characteristics to some specified degree. These characteristics need not, themselves, be normative. If this is true, then being a good X can be translated without loss to being an X with those non-normative features or characteristics specified by the kind of thing X is, to an appropriately specified degree. Therefore, goodness can be understood as being identical (in some sense) to being for X (Alexander 2012, 91-95).

The second part of the first argument needs to show the other implication, that being is identical (in some sense) to good. Remember that on the attributive account, something is good insofar as it fulfills its function or exemplifies its nature to a specified degree. Many kinds of things in the world go through stages of development toward their most full form. While this argument is not as clear in the case of artifacts

119 and inanimate objects, a great many living things pass through several developmental stages on their way to becoming a mature, adult member of their species. For our purposes, let us imagine a first stage of the development of a thing. Alexander suggests that we imagine this as follows. “Now, at stage 1, X is the barest sort of K [or member of its kind] possible. That is, at stage 1, X is an unqualified K. X is, as it were, a mere

K,” (Alexander 2012, 95). When we define goodness as having some characteristics to some degree related to the kind of thing X is, we can also break down our account of goodness by applying our judgments to entities at relative stages of their development.

An acorn is not a full-fledged oak tree, but one acorn can be better than another with respect to its having characteristic features of that early stage of its development.

Insofar as an X is a member of its kind, even at stage 1, it possesses those properties that are good at stage 1. So X is good, qualified to its status as a stage 1 X. This just means, though, that X’s having some set of characteristics (whatever those are for the kind of thing we pick out) to some degree is identical to X’s being good. So X’s being is identical (in some sense) to X’s being good.

This first argument should not sound entirely new to readers of the chapter. In fact, it is an updated version of Aquinas’ own arguments (and the arguments of his later interpreters) that attempt to show how being and goodness are convertible on the Nature

Approach (see c-d). Although Alexander is working in a different framework, he notes at several places that his project has affinities with both Aristotelian and Thomist metaphysics.45 Similarly, his own arguments (given above) to connect CT and the attributive account of good use much of the same language, same concepts, and roughly

45 Most of chapter 4 in Alexander 2012 is focused on defending an Aristotelian-style account of function, and Chapter 5 in that same text at several points mentions the sympathies the project has with Aquinas and CT. 120 the same argumentative moves as the arguments Aquinas gives to connect being and goodness in the Summa Theologiae and the Summa Contra Gentiles, as shown above.

The second argument that connects the attributive account of goodness and the convertibility theory does so indirectly. The attributive account not only treats “good” as an attributive property, but also treats “bad” or “evil” as an attributive property. So on the attributive account, just like nothing is just “good” irrespective of the kind of thing it is, but only good relative to its nature or the kind of thing it is, so nothing is just

“bad” or “evil” irrespective of the kind of thing it is, but only bad or evil relative to its nature or the kind of thing it is. If being virtuous is a good thing for a human being, based on the kind of thing she is, then to the degree she lacks virtues she will to that degree be a bad or evil human being. If having appropriate emotional responsiveness is a unique characteristic of human nature, then this lack is bad or is an evil, the degree of which depends on the severity of the lack. On this view, a completely evil thing could not exist, because only if evil or bad were predicative properties could something be labeled completely or wholly bad or evil. Since the attributive account denies this, it denies that anything completely evil could exist. Further, on the attributive account evil is parasitic upon an already existent notion of goodness. Evil is some lack or absence

(or something that causes a lack or absence) of a characteristic or a feature fundamental to some thing’s pursuit of its fulfilled nature. Without already having a sense of what a thing’s nature is, and what it would be for that thing to fulfill its nature (that is, what would count as good for that thing, or a good thing of that kind), we cannot talk about evil as a lack or absence of a characteristic or feature, or as failing to fulfill some important part of the nature of the thing under consideration. These results essentially

121 spell out PE. An extremely simple version of PE is that evil is an absence or lack of some kind, where that lack ought not to be there (this is a quick and rough statement of the theory, but the next section explores PE in some detail, so this characterization will suffice for now). An implication of PE is that nothing could be completely or utterly evil, as for it to exist or be at all it is evincing some goodness. Another implication is that evil is parasitic on goodness. There could be a universe of all and only good things, but there cannot exist an instance of evil without some good existing.

Alexander argues that if the attributive account of good and evil is true, then PE is true. He then states that PE and CT are related conditionally. If the attributive account of good and evil implies PE, and if PE implies CT, then the attributive account of good and evil implies CT. Accordingly, if CT implies PE, and PE implies the attributive account of good and evil, then CT will imply the attributive account of good and evil.46

Section V. The Privation Theory of Evil

The PE was once a commonly held corollary of CT. In contemporary circles, there are few defenders of a PE, and most of those defenders are neo-Thomists or

Christian philosophers influenced by scholastic thought. Our purpose in this first sub- section will be to explain PE as it relates to the Nature Approach of CT. To some degree, this will depart from Alexander’s treatment of the privation theory (in

Alexander 2012), but it will give the reader a general sense of the privation theory as both a historical artifact and a contemporary, though rare, account of the nature of evil.

Just as CT’s account of goodness must encompass both a metaphysical and moral

46 Alexander does not go this far, but historically many philosophers thought that CT is what implies PE, not vice versa. Accordingly, I will take it that CT and PE are related biconditionally, though this is contentious (see Midgely 1984 and Larrimore 2011 for opposing accounts). 122 account of goodness, so PE’s account of evil must encompass both the metaphysical and moral aspects of evil. It will seem strange in some cases to see the metaphysical gaps labeled as evil, but this should strike the reader as not much weirder than the metaphysical fullness of objects being counted as good. In the second sub-section, I will argue against the very common objection against PE that pain is a counterexample to any PE. If pain cannot be accounted for in terms of privation, and if pain is an evil, then any version of PE that takes pain to be a real-world phenomenon will fail.47

The main reason to discuss PE in this dissertation is primarily negative. I believe that one of the main objections to CT is its relationship to PE – CT entails PE.

Most contemporary philosophers hold PE to be obviously false, for the reasons given above or for others considered independently (as cataloged by Kane 1980 and Calder

2007). I remind the reader that my aim is to defend what I've referred to as the

"reflective plausibility" of CT. What I mean by this, as explained earlier in the chapter, is that not only is CT is not obviously false, as many have assumed, but it is plausible on the metaphysics supported above. Since PE is taken to be obviously false by most philosophers, and since PE and CT are biconditionally related, it is no great leap to assume that CT is taken to be obviously false by most philosophers. In the second sub- section, I will try to persuade the reader that PE has a certain level of plausibility and that the main objection offered to it is nowhere near decisive. If I succeed, I will have shown that the most damning obstacle to positively judging CT’s plausibility is itself bereft of needed support.

47 This objection is pressed by Kane 1980 and Calder 2007 and responded to in kind by Anglin and Goetz 1982, and Alexander 2012. For a defense of PE on different grounds and a critique of other criticisms offered by Calder 2007, see Grant 2015. For a schema regarding how PE's must account for pain, see Swenson 2009. 123

V.A Explaining PE

For many of the scholastic philosophers, good and evil were seen as opposites.

This led a young Augustine to be impressed by a dualist, Manichean ontology of good and evil. According to this view, good and evil are both positive qualities in the world, and each is produced by separate but powerful good and evil deities struggling for dominion over the world. In many ways, this makes sense of the nature and variety of both goodness and evil in our world. However, Augustine was later influenced by neo-

Platonic Christian thought, according to which evil is not some quality in-and-of-itself, but rather a lack or absence or privation in the world. Augustine did not work this out as thoroughly as some later scholastics, but his writings have had significant influence in the understanding and debate regarding PE. Numerous other scholastic writers, such as Thomas Aquinas and Francisco Suarez, worked out highly developed accounts of PE, including considering counter-intuitive or problematic aspects of the doctrine (see

Aquinas 2001 and Gracia 1991). Since the end of the middle Ages, privation theories of evil have been by and large forgotten or discarded for other accounts of the nature of evil.48 Yet, recently PE has been aptly defended by a number of contemporary philosophers as shown throughout the remainder of this section.

Just as there are two ways to think about CT of being and goodness, there are also two ways to think about PE. On the Participation Approach, where a thing receives its goodness by virtue of its participation in God (or God’s substantial creative sustaining of that thing), the story about evil as a privation is simple. A thing or attribute is evil insofar as it fails to participate in (or be supported by) God’s goodness.

48 For a recent example, Kane writes “Aside from orthodox Thomists, few philosophers, either theists or their critics, accept this theory any more,” in Kane 1980, 43. 124

But I will focus on the Nature Approach to PE for two reasons. First, just as the

Participation Approach to CT presupposes the existence of God and is thereby unhelpful in supporting an argument for the existence of God, the Participation

Approach to PE does the same. The second reason to focus on the Nature Approach of the privation theory is that the Participation Approach to PE is very opaque in its account of evil. What, exactly, does it mean that a thing fails to participate in, or be supported by, God’s goodness? How would it be the fault of a person if God has chosen not to add his substance to that thing, in moral terms? How does this shed any light on moral responsibility, or even on what we might call natural evils?49 Since the approach is obscure as applied to analyzing both moral and metaphysical evils

(including natural evils), as well as presuming a proposition we cannot take for granted in the context of an argument, we will abandon the Participation Approach to evil as a privation, and aim at explaining a PE drawn from and consistent with the Nature

Approach to CT.

How does one treat goodness and evil as opposites without admitting evil as a positive quality of the world? Since CT states that being and goodness are the same in extension but different in intension, then everything, insofar as it is a being is good.

This looks puzzling, because our initial inclination is to read this as “Evil does not exist,” or “There are no evil existent things.” However, PE does not amount to the claim that there are no evils in the world, or that evils are somehow less important due to their not being a positive quality, but instead is about the nature of evils we see in the world. Evils are explainable as a privation or absence or lack of some quality or

49 For purposes of clarification, I do not mean “natural evils” as the terminology is often used in the literature relating to the Problem of Evil. Instead, I mean “natural evils” or “natural badness” as that sub- type of metaphysical evil or badness separate from both moral and artificial evil or badness. 125 arrangement. This needs to be further modified in order to understand the theory and to give it legs. Evil is not just a lack or an absence of anything relevant to any other thing, but a privation or a lack of quality that ought to be in the fulfilled nature of a thing. It would be strange to consider my lack of compound vision or replayability an evil for me, though these things might be evils as they afflict other kinds of things.50 It would make sense to consider a lack of vision at all an evil for me, or a lack of mobility since these are qualities required by the most mature version of my nature.51

We can also show that this holds true for non-physical (or perhaps, non-bodied) qualities characteristic of my nature. Lust or greed is an evil because of a disorder in my valuing those things I ought to value, and their relative importance in my evaluative scheme. This disorder can be analyzed as a lack or an absence of both the appropriate ordering in my evaluative scheme, as well as a lack of importance assigned to values that ought to be higher in my hierarchy of values. Note also that this need not presuppose anything over and above a naturalistic metaphysics. As stated earlier in the chapter, naturalistic virtue theory accounts not only posit goods for human beings based on their natures, but also evils for human beings based on their natures. We can also use this analysis with regard to other animals and objects in nature, without requiring any designer (or in the case of artifacts, any designer other than human beings).

Something is bad for an elephant, or an oak tree, or a computer, insofar as it prevents

50 Some insects and some games, respectively. 51 This is perhaps the most ethically contentious part of CT and PE, given what it would imply about various kinds of disability in general. I cannot tackle this difficult and sensitive issue here – I only mention it to note that this is an important implication that should be dealt with head-on in any full-blown defense of CT. 126 them from fulfilling their nature, or causes an absence of a quality they ought to possess, for the complete fulfillment of their nature.52,53

On the privation theory, every evil can be characterized as an absence or a lack, not in general, but with regard to the nature of the thing in question. This is true of characterizing how evils affect individuals, as well as how evils are constituted. This also shows how we can characterize certain frustrations of metaphysical goodness being fulfilled as metaphysical evils for the object or entity in question. To show this, let us explore three cases.

A. In one circumstance, we have a person whose leg is completely paralyzed and numb. In another circumstance, we have a person whose leg is in terrible pain. In both cases, the subject’s leg is unusable for locomotive purposes or sensory response.54 B. A person desires to torture and kill another. That second person is the subject of being tortured, and then their life is ended.55 C. A sapling oak is consumed by a fire. As it is consumed, a person rushes in with a pail of water and puts out the fire.56

In case A, we have a favorite foil for the opponents of PE. Pain, it is said, is not simply an absence (as paralysis or numbness might be) but as a positive, felt quality over and above a lack of pleasure. The proponent of the privation theory, though, can

52 Is this version of PE, based on the Nature Approach, more illuminating than that of PE based on the Participation Approach? There might be thought to be such a significant gap between natural goodness/badness and moral goodness/badness that using the Nature Approach does not help. I think this is a good question, but one that I think depends on one’s appraisal of the corresponding CT. If the Nature Approach of CT is correct, then metaphysical goodness will include natural goodness, artificial goodness, and moral goodness. To the extent that people find this treatment implausible for goodness, they will likely find it to be implausible for badness or evil as well. I thank Neal Judisch for this point. 53 Again, I remind the reader that if one is skeptical about objects having natures, then one will be skeptical about the line of reasoning given. Note, though, that this is a general problem with any account that involves natures of things to any significant degree, and not a specific problem with CT or the privation theory of evil. 54 Comparing pain and the absence of pleasure or sensation goes back to at least Aquinas and Suarez, as discussed in Aquinas 2001 and Gracia 1991, but contemporary writing on the topic include cases specifically comparing pain and paralysis or absence of feeling, such as Alexander 2012, Calder 2007, Goetz and Anglin 1982, and Kane 1980. 55 Analyzing both the evils of vicious traits, as well as the evils of the actions of vicious people is also a longstanding feature of the privation theory of evil, in both historical and contemporary writings. 56 Although the metaphysical usage of the term evil has fallen out of favor for a strict moral conception, it can be found in scholastic writers. 127 analyze both instances given in case A as absences. The person whose leg is paralyzed or numb suffers from a lack of motion and sensory perception in the leg. It is an evil for that person, and it consists in a lack of those two qualities which a fully function human being ought to have. The person whose leg is in terrible pain is also on the receiving end of an evil. The pain, as it causes both an inability to use the leg or feel normal sensory perceptions in the leg constitutes an evil. We might judge that the person whose leg is in pain is on the receiving end of a greater evil, but we could still appeal to the privation theory by showing that in the case of severe pain, several other qualities needed for the full function of the person (ability to maintain cognitive focus, a sensitivity to important factors around them, etc.) are lacking because of the intense pain in the leg. In both cases, the privation theory can adequately analyze the evil occurring. I will return to the case of pain below.

In case B, there are several evils available to analyze. First, there is the evil which consists of the character and characteristics of the sadistic torturer/murderer themselves. Second, there is the evil which consists in the effects the torture and murder have on the victim. In both cases, the privation theorist can show that the evil consists in privation. As regards the torturer, what makes this person's beliefs, desires, and tendencies evil is their attachment of importance on fulfilling their own desires by seeing another suffer and be extinguished, and the lack of order shown in their disordered set of values. If human beings, by their nature, are moral and social beings, then something important is missing in a torturer, something important that ought to be there to fulfill the human beings nature. For the tortured and eventually murdered, a host of evils occur. The pain, the suffering, and the end of their lives. The pain they

128 experience can, as before, be shown to result in the absence of a number of other goods for them (tranquility, contentment, ability to maintain cognitive focus, etc.). Their suffering can be characterized as depriving them of various aims they take their life to have, as well as a frustration or impediment to their vision of their life. The end of their life, of course, is an evil precisely because it results in their complete lack of being, and all of the goods associating with existing.

Case C is perhaps the strangest of the three cases. For most contemporary philosophers, it is bizarre to talk about evil in relation to trees burning down or water extinguishing fire. The important thing for our current purposes is not to get involved in a debate about the appropriateness of a broad conception of evil which includes both metaphysical and moral characteristics (as does our earlier conception of goodness).

Instead, we focus on how an account of evil as a privation of a quality central to a thing’s nature can cover even these strange, counterintuitive cases. If an oak sapling’s nature, realized at its fullest level, is to grow into a sturdy, old oak tree, then its being burned down can be counted as an evil. It is the destruction, the rendering into nothing and thus preventing the completion of the oak tree’s purpose, which is an evil here.

Similarly, insofar as it is in the nature of fire to consume, dousing a fire with water prevents this, and in fact renders fire nonexistent, the dousing of a fire is an evil, at least with regards to the fire. There are numerous difficult questions here. How do we understand the idea that immobile, insensate living things can have a nature, much less that their natures can be frustrated or flouted? How do we understand a process, such as fire, as having a nature at all? In what way does it make sense to speak of an evil done to a tree or to a fire? In fact, one might regard the entirety of case C as nonsense. Oaks

129 and fires don’t have natures to be frustrated, an opponent might say, and stretching the term evil to cover such frustrations is not really recognizable as evil at all. I think that this can be right and still not bother the privation theorist. All the privation theorist needs to hold is that if a thing has a nature, and if that nature has qualities that can be lacked, resulting in the denial or incompleteness of the full fruition of that things nature, then it can be analyzed using PE. If it turns out that oak trees and fires do not have natures, then this will not bother the privation theorist. If, on the other hand, the natural entities and processes do have natures, PE can be used to analyze them in metaphysical, as opposed to moral, terms.

V.B Defending PE

In this sub-section, I defend the viability of CT by arguing against the most famous and obvious objection to PE – the objection from pain. It is commonly assumed that the badness of pain is, without contention, some positive, felt quality that cannot be suitably analyzed as a privation. If this is the case, and if PE requires us to deny this or cannot account for it, then that will act as a reductio of PE and consequently of CT. In line with my earlier remarks about "reflective plausibility," my primary objective in this sub-section will simply be to throw a wrench in the gears of the obviousness and/or decisiveness of this favored line of critique. Here is the structure of what I imagine the argument against the reflective plausibility of CT to be, from the objection from pain.

1. If PE is false, so is CT. 2. Pain is a counterexample to PE – pain is both an evil and not a privation. 3. CT is false.

If this is a suitable construal of the implicit reasoning behind casually disregarding CT, and I believe that it is, then providing reasons to doubt 2 and to

130 support the idea that pain is not a privation will serve to reduce our confidence in 3.

Taking on this task will bolster CT by providing a defense for what its critics take to be its weakest and most fatally flawed commitment.

The idea that pain is obviously evil in a way that goes beyond any absence or lack is not only found in scholars wrestling with theology or phenomenology or metaethics – instead, it is available and immediately apparent to all of us on the basis of little introspection. Kane writes

The difficulty is that pain seems clearly to be more than merely the absence of its contrary opposite. There is a marked difference between a limb which merely lacks feeling - is numb or paralyzed or anesthetized - and one which is racked with pain. In the former case it is quite plausible to say that there is merely a privation of something, namely normal feeling, that under usual circumstances would belong to the limb. But it is clearly inadequate to describe a limb aching with pain as suffering merely a privation of good health or normal feeling. When pain occurs in the body, there is something new and different in a person's experience which is not present when the body has simply lost feeling. (Kane 1980, 49)

Similarly, Calder writes

… pain is not simply the absence of feeling or pleasure, it is a positively bad sensation or feeling. The absence of feeling or pleasure is numbness or experiential paralysis; it is akin to the way a body part feels under local anesthetic. Pain, on the other hand, is a felt quality, one that we typically try to avoid. (Calder 2007, 373)

Alexander, framing the pain objection before offering a response, writes

The badness of a pain is something over and above the badness that a pain may indicate (say, the badness that brings about the pain). “The pain is caused by the absence, it does not consist in it.” To repeat, the badness of a pain is the feeling of it; its badness lies in its phenomenology, and its phenomenology is essential to it or at least intrinsic to it. Thus, pains are essentially, or at least intrinsically, bad. (Alexander 2012, 102).

Focusing solely on PE at this point, the objection from pain would go as follows.

131

4. If PE is true, then all evils are privations. (Definition of PE) 5. Pain is an evil. (Common Assent) 6. Pains are not privations. (Common Assent) 7. Therefore, PE is false. (4-6)

Premises 4 and 5 are fairly straightforward. 4 is simply what PE says about the nature of evil. 5 seems deeply rooted in our everyday experience. 6 is also, as noted above, seemingly obvious. However, it is important to examine exactly what might be meant by 6, and spelling this out will help us to determine exactly how a proponent of PE might respond.

Calder clearly has in mind the idea that painfulness is a qualitative conscious experience, as he notes in Calder 2007, endnote 11. Kane, too, appears to appeal to the positive, felt quality of pain in demonstrating that it is inconsistent with analyzing pain as a privation. In both of these cases, the support for 6 is something like

8. All sensations are positive, felt qualities. 9. Pains are intrinsically or essentially sensations. 10. Nothing that is a positive, felt quality can be a privation. 11. Pains are not privations.

At this point the question is whether or not the foregoing argument is true – is 9, the claim that “pains are intrinsically or essentially sensations” true? To answer this, we would need to examine some contemporary accounts of the nature of pain.

Alexander takes exactly this line of argument. He shows that PE is consistent with three contemporary accounts of pain; Michael Tye’s strong representationalism about pain in “Another Look at Representationalism About Pain” (Tye 2005), Klein’s imperative account of pain from “An Imperative Theory of Pain” (Klein 2007) and further developed in “What Pain Asymbolia Really Shows” (Klein 2015), and Austen

Clark’s view that the evil or badness of painfulness is not a sensation but an attitude

132 towards that sensation, supported by the phenomenon of pain asymbolia57 (cases where one has a sensation of pain but does not negatively evaluate it) in “Painfulness is Not a

Quale” (Clark 2005). In each case, Alexander argues that the account of pain as a privation given by PE is consistent with some other contemporary account (as an indicator or representation of some lack of bodily integrity from Tye 2005, or as a command that some state or action cease from Klein 2007, or as an affective judgment about the badness of pain due to a lack of some positive internal state or external bodily integrity from Clark 2005). I will not rehearse these arguments and interpretations here

– I bring this to the reader’s attention primarily to show that the seemingly intuitive fact that pain must be essentially bad due to its qualitative content is far from a settled matter. If this is correct, then PE (and consequently CT) may retain its status of reflective plausibility.

I want to give a further argument against 9 based on the work done by Don

Gustafson in “Categorizing Pain” (Gustafson 2005). In it, he gives a short history of the concept of pain and the role it historically played in various societies, cultures, and even scientific accounts. In some cases, pain was conceived of as external, as an issue of the body and not a sensation. In other cases, pain was valuable precisely because of its aversive character. Gustafson argues convincingly that for reasons both scientific and historical, we should reject the idea that all pains are intrinsically a simple sensation of one type. Pain is a complex phenomenon with a variety of causes, involving the body as well as the mind, involving some motivational content as well as affective components, and true to such a varied phenomenon it has, at different times and in

57 For a competing account of how we ought to analyze pain asymbolia, see the aforementioned Klein 2015. 133 different cultures, been conceived of in competing and even opposing ways.

Gustafson's closing remarks regarding the recent history of pain concepts are telling against the seemingly obvious nature of the qualitative character of pain.

The conception of wholly subjective objects of introspection became a philosophical staple. Pain, in turn, became the paradigm subjective object of secure introspection. Incorrigibility of pain reports and expressions attained the status of a dogma. This is my story of the internalization of pain. It is a history. Its episodes are contingent events. Our conception of pain is delivered to us by historical, scientific, and philosophical contingencies. If this is plausible, it is also plausible that we may wish to reconfigure our conception of pain or the categorization of pain if we find another more useful to current interests. After all, if the history just now sketched is possible, then there is no necessity in the idea of pain as an essentially hidden or inner subjective condition, accessible only by means of self-consciousness. (Gustafson 2005, 231)

If this is true, we have reason to doubt the truth of 9. It may be that pains are sensations, but if our concept of pain, and how we categorize and characterize it, is as fragile and historically dependent as Gustafson argues it is, then this gives us some reason to doubt the seemingly obvious account of pain as essentially a qualitative state.

I don’t take either of these considerations to compel assent to the conclusion that

PE can account for pains – rather, I note them in order to show that a) the nature and character of pain is itself a hotly contested issue that is far from obvious, and b) that some of these contemporary theoretical accounts of pain are either consistent with PE

(Alexander 2012) or give reason enough to serve to undermine an important premise in the objection from pain (Gustafson 2005). For these reasons, CT can be said to be reflectively plausible – it is not epistemically responsible to casually disregard the plausibility of the theory on the basis of obvious pain considerations.58

58 An additional consideration in favor of CT by route of PE is PE's value beyond any theological projects. Grant 2015 attempts to show how this kind of defense might go by not only objecting to several critiques against PE, but also by arguing that PE best accounts for evil in the moral sense. Midgely 1984 134

Section VI. Summary

In section I, I introduced the chapter’s structure as a whole. In section II, I gave a short historical background to CT. This is important because it gives us a sense of the importance of this theory in earlier philosophical epochs and it shows that there are hundreds of years of development behind CT. In section III, I examined various arguments given by Aquinas for CT. This is important because Aquinas’ version of CT is the most well-developed and worked-out version of CT. It is also an example of the

Nature Approach to CT, which I have argued is important for the dialectical purposes of this dissertations. In section IV, I explained the Nature Approach to CT in detail and demonstrated how it is possible to meaningfully and coherently talk about artificial objects, natural non-human entities, and human beings in terms of this CT approach. I also demonstrated how it is possible to talk about both metaphysical and moral goodness and being. In addition, I demonstrated how it is possible to talk about degrees of being and degrees of existence in the same way as it is possible to talk about degrees of goodness. The last part of this section gave an example of a current metaethical theory that is consistent with and well-supported by CT. In section V, I analyzed PE and explained how it is possible to meaningfully and coherently talk about pain, immoral actions, and the destruction of non-conscious objects in terms of PE. I also defended the reflective plausibility of CT by undermining the confidence critics should have in the central objection to PE. In the next chapter, I argue that all ontological arguments that hold that there is an important connection between goodness and

similarly suggests that PE is the only viable account of evil – there is little to recommend a kind of Manichean, positive account of evil given our nature. If Grant and Midgely are correct, then adopting CT will provide all of the advantages of adopting PE, in addition to those advantages I argue for in the final chapter of this dissertation. Finally, a defender of PE could simply deny that pain is an evil, though this would itself be a significant undertaking. 135 existence or being are practically committed to CT. I further argue that CT as a theory has a number of other underappreciated advantages as a metaethical theory.

136

Chapter 5: The Central Argument

Section I. Introduction

The central thesis of this dissertation is that many historical and contemporary versions of the ontological argument depend upon the Convertibility Thesis (CT) for their success. In order to persuade the reader, I break down this argument into five sections. In section II, I will show that a number of historical and contemporary ontological arguments contain a premise that links goodness or greatness or perfection to existence or necessary existence (hereafter referred to as the Linkage Premise). I will do this by enumerating the premises or sections in each of these arguments that I take to involve this tight connection between greatness/goodness/perfection and existence. In section III, I will show that in each case, very little or no argument is given in support or justification of this premise, or the argument involves either CT or a close analog. In some cases, no support for the Linkage Premise is given because the premise involves stipulative definition. If the argument already presupposes CT or some close analogy, I will not need to go further. If the argument does not, I will demonstrate this by close textual examination. In section IV, I will show that CT entails the tight connection required for the success of the argument. If CT is true, not only are the various premises linking greatness/goodness/perfection to existence true, but they are justified by reference to a powerful and well-worked out theoretical background. In section V, I will show that, in the absence of competitor theories, CT is the theory most well suited to support the Linkage Premises. I will also demonstrate how CT rules out the AntiGod parody objection, which I have shown in chapter four to be the most powerful objection to ontological arguments. In section VI, I explain the conclusion of the argument,

137 which is that any person committed to the viability of the ontological argument (of a certain form) is at least practically committed to the viability of CT.

In standard form, the argument would go as follows.

1. Numerous versions of the ontological argument, both historical and contemporary, rely upon a Linkage Premise, that is, a premise that there is some tight connection between goodness, or greatness, or perfection, and existence (either existence in general or necessary existence). 2. In the case of many of these arguments, very little or no argument or justification is given for this Linkage Premise. In other of these arguments, the Convertibility Thesis of Goodness and Being (hereafter, CT) or something close to it is already explicitly or implicitly required 3. CT entails the Linkage Premise (if CT is true, then the Linkage Premise will be true). 4. In the absence of competitor theories, CT is the most well-suited and detailed theory to adopt in order to substantiate the truth of the Linkage Premise. 5. Therefore, proponents of any ontological argument that both a) relies upon a Linkage Premise and b) does not include a competitor theory are at least practically committed to the viability of CT. Section II. The Linkage Premise

Although it is possible to eliminate the language of greatness, perfection, or maximality from ontological arguments, this is rare.59 When we examine arguments offered in the literature most do not directly connect existence or necessary existence to

God without any intermediary. Instead, notions of goodness, greatness, perfection, or maximality are used to connect existence or necessary existence and God. If existence and goodness (or the other relevant notions) are tightly connected somehow, and God is

59 For example, the following are all possible ontological arguments given in Oppy 1995. 1. It is possible that it is actually the case that God exists. (premise) 2. Therefore, God exists. (from 1) or 1. I conceive of an existent God. (premise) 2. Therefore, God exists. (from 1) or 1. By definition, God is an existent being. (premise) 2. Therefore, God exists. (from 1) 138 defined as the most perfect being/the greatest conceivable being/the instantiator of maximal greatness, then it will naturally follow that God exists or exists necessarily.

Of course, matters are a bit more complicated than that – if we are appraising the argument we can still reject the Linkage Premise itself, or the possibility premise, or that S5 (or Brouwer, or whichever logical system in use) is the correct modal logic for metaphysics, among other moves. But one thing we will not be able to object to if we have a Linkage Premise is the simple connection between God and existence, as there is no longer any simple, direct connection to be had. Instead, it is the Linkage Premise that allows us to connect God and existence (or necessary existence). We do not suggest a brute connection between existence and God as is done in the versions of the argument given above, in part because this move seems objectionable even to many theists. We do not attempt to define God into existence by simply adding in “existence” to the concept as Kant and others have charged. Instead, we draw on the deep-seated intuition that existence and goodness/greatness/perfection have some type of tight conceptual relationship, spelled out differently in different authors. This would help the theist’s case. If there is some connection between goodness/greatness/perfection and existence (or being, or necessary existence), and if the concept of God has long been understood to contain some description such as “the greatest conceivable being, the most perfect possible being, the ens perfectissimum, the maximally greatest being, etc.” then the argument flows more naturally – instead of a jarring attempt to conjoin necessary existence to God through brute stipulative force, the inclusion of existence or necessary existence in the concept of God is due to the natural and agreeable connection between it and the concepts of goodness/greatness/perfection.

139

Because of the preceding considerations, the most natural starting point for our investigation is the premise in various ontological arguments that is supposed to link existence or necessary existence to God by way of invoking greatness or perfection. I list several examples below, and then examine them further in the following paragraphs.

1. “For if it were only in the understanding, it could be thought to exist also in reality—something which is greater [than existing only in the understanding].” (Anselm 2000a, 94) 2. “For there can be thought to exist something which cannot be thought not to exist; and this thing is greater than that which can be thought not to exist.” (Anselm 2000a, 94) 3. “So it is no less repugnant to think of a God (that is, a supremely perfect being) lacking existence (that is, lacking some perfection) than it is to think of a mountain lacking a valley.” (Descartes 1980, 86) 4. “Then it is plausible to suppose that the maximal degree of greatness entails maximal excellence in every world. A being, then, has the maximal degree of greatness in a given world W only if it has maximal excellence in every possible world. But maximal excellence entails omniscience, omnipotence, and moral perfection. That is to say, a being B has maximal excellence in a world W only if B has omniscience, omnipotence, and moral perfection in W -- only if B would have been omniscient, omnipotent, and morally perfect if W had been actual.” (Plantinga 1977, 108) 5. “Necessary existence is positive. Positive means positive in the moral aesthetic sense (independently of the accidental structure of the world)... It may also mean pure attribution as opposed to privation (or containing privation).” (Sobel, 2003, 145) 6. “Previously, I rejected existence as a perfection. Anselm is maintaining in the remarks last quoted, not that existence is a perfection, but that the logical impossibility of nonexistence is a perfection. In other words, necessary existence is a perfection.” (Malcolm 1960, 41) 7. “Think of a perfection (P2) as a property that it is necessarily better to have than not; and define the property of being supreme (S1) as the property that a thing has if and only if it is impossible for something to be greater and impossible for there to be something else than which it is not greater.” (Maydole 2009, 580) 8. “(1) God is a being having all perfections. (Definition). (2) perfection is a simple and absolute property. (Definition). (3) Existence is a perfection. (4) If existence is part of the essence of a thing, then it is a necessary being. (5) If it is possible for a necessary being to exist, then a necessary being does exist. (6) It is possible for a being to have all perfections. (7) Therefore, a necessary being (God) does exist.” (Look 2013, section 7.1)

140

Examples 1, 2, 4, and 6 require that greatness is linked to either existence or necessary existence, that a thing existing or necessarily existing is greater than it failing to exist.60 Examples 3, 5, 7, 8 and 9, require that perfection is linked to either existence or necessary existence, claiming either that existence or necessary existence is a perfection, or that the most perfect thing must exist.

II.A Anselm

Anselm’s argument occurs in the context of faith seeking understanding. For him, God is “that than which nothing greater can be thought,” (Anselm 2000a, 93). In the Proslogion 2, Anselm provides a clear and specific link between greatness and existence when he states of the idea of God that “For if it were only in the understanding, it could be thought to exist also in reality—something which is greater

[than existing only in the understanding],” (Anselm 2000a, 94). Here the distinction is between existence-in-the-understanding and existence-in-reality, but we need not rely on this distinction to get Anselm’s point across. If an object, greater than which none can be conceived, can be merely understood (that is conceived, but fail to exist), we could argue that in the act of understanding it we would conclude that it exists, for the reasons or similar reasons that Anselm gives. What we can generalize from this is some principle such as “For a thing to exist is greater than for that same thing not to exist” or

“Existence is a great-making property.” Anselm gives a similar argument in the

Proslogion 3 but with regard to necessary existence. Since both of these arguments

60 Although Gödel’s argument only speaks of positive properties, the definition of a positive property is that the property is positive in “the moral aesthetic sense,” or “in the sense of pure attribution” so I think this is a suitable interpretation of his work, as I argue in a later section. 141 show a tight link between greatness and existence (or necessary existence) they serve our purpose in this stage of the argument.

II.B Descartes

Descartes explicitly connects existence with perfection in his version of the argument. For Descartes, God is defined as a supremely perfect being, a being that by its nature cannot lack any perfection. So, Descartes shows that God exists by defining existence as a perfection. After explaining how he comes to have an idea of God, in his own experience, Descartes says “So it is no less repugnant to think of a God (that is, a supremely perfect being) lacking existence (that is, lacking some perfection) than it is to think of a mountain lacking a valley,” (Descartes 1980, 86). Here, existence is defined as a perfection, thus providing the linkage between existence and perfection.

II.C Plantinga

Plantinga’s discussion also relies on an explicit connection between greatness and existence. In an initial discussion of the ontological argument and how one might modalize it, he says “But now suppose we think a bit more about the being than which it is not possible to be greater. This being possesses a maximal degree of greatness; a degree of greatness that is nowhere excelled. That is to say, its greatness is not exceeded by the greatness of any being in any possible world,” (Plantinga 1974, 214).

This is one point where he explains what maximal greatness might mean, and it seems to be a pretty straightforward development of Anselm’s own vocabulary. Later, he argues that God must have those properties that make him an object worthy of worship

(moral goodness, power, intelligence) essentially – it is not enough for the theist that he

142 have them in this world but is incompetent or stupid in another. In discussing this concept of God, he makes the distinction between greatness and excellence as follows.

Those who worship God do not think of him as a being that happens to be of surpassing excellence in this world but who in some other worlds is powerless or uninformed or of dubious moral character. We might make a distinction here between greatness and excellence; we might say that the excellence of a being in a given world W depends only upon its (non world-indexed) properties in W, while its greatness in W depends not merely upon its excellence in W but also upon its excellence in other worlds. (Plantinga 1974, 214) Plantinga goes on to argue (and I think, accurately) that if we think of greatness this way, perhaps we can get rid of the supposition that existence is not a perfection.

However, by utilizing this definition, he has explicitly connected degrees of greatness to the possession of excellence in other possible worlds (and a precondition of having any properties, as he notes, is existing to have those properties in the first place).

Plantinga’s definition of maximal greatness does require necessary existence, after all.

II.D Gödel

Gödel’s own ontological argument is one of the most spare of those on offer. It takes up only 1-2 pages in his notebooks. It takes the form, for the most part, of logical notation with some very brief notes giving definitions of key terms. The most we get about positive properties, which are instrumental to the proof, are as follows.

“Necessary existence is positive. Positive means positive in the moral aesthetic sense

(independently of the accidental structure of the world)... It may also mean pure attribution as opposed to privation (or containing privation),” (Sobel 2003, 145). This is not entirely clear (as others have noted) but it does suggest a connection between goodness or greatness (a gloss on the moral-aesthetic sense of a positive property) and necessary existence.

143

II.E Malcolm

Malcolm’s argument is, as in some previous cases, a development of Anselm’s own version of the argument. In trying to explain Anselm’s argument in Proslogion 3,

Malcolm says that

Previously, I rejected existence as a perfection. Anselm is maintaining in the remarks last quoted, not that existence is a perfection, but that the logical impossibility of nonexistence is a perfection. In other words, necessary existence is a perfection. His first ontological proof uses the principle that a thing is greater if it exists than if it does not exist. His second proof employs the different principle that a thing is greater if it necessarily exists than if it does not necessarily exist. (Malcolm 1960, 46) Many interpreters of Malcolm have suggested that he relies upon a notion of

God’s being unlimited to link it to his eternal existence, as opposed to linking greatness or perfection to existence of any kind. He does give this argument in the article quoted, but in his own discussion the notion of God’s being unlimited includes the notion of greatness. Again, he writes “God is usually conceived of as an unlimited being. He is conceived of as a being that could not be limited, that is, as an absolutely unlimited being. This is no less than to conceive of him as something greater than which cannot be conceived,” (Malcolm 1960, 47). For Malcolm, then, an absolutely unlimited being just is a being of with the highest possible degree of greatness.

II.F Maydole

Maydole’s Linkage Premise occurs in the form of a stipulative definition, and is quite a bit different from others. In fact, Maydole links perfection to existence not directly, but through the notion of supremacy. A being being supreme, and supremacy being a perfection results in the beings’ existence. In fact, the Linkage Premise listed

144 above for Maydole’s argument is perhaps the least clear of all of them, since it is not direct (i.e. the linkage between perfection and existence requires an intermediary concept, that of supremacy). That being said, it seems to me that a crucial part of

Maydole’s argument requires a relationship between existence and perfection, even if it occurs at a slight remove.

II.G Leibniz

Leibniz is yet another writer who explicitly makes the link between existence and perfection. Defining perfection, he writes that “by a perfection I mean every simple quality which is positive and absolute or which expresses whatever it expresses without any limits,” (Leibniz 1970, 167). Leibniz then sets out to show that all perfections are compatible with one another, in order to demonstrate that God, defined as “a subject of all perfections or a most perfect being,” is possible. The most relevant part of this discussion is at the end, where Leibniz says “Hence it is clear that this being exists, since existence is contained in the number of perfections,” (Leibniz 1970, 168).

Existence is a perfection, according to Leibniz.

In this section, I have demonstrated that each of our authors is committed to some version of the Linkage Premise. In the following section, I show how such a premise is justified for each of our authors.

Section III. Justification or Support for the Linkage Premise

I have shown that in the case of each of the arguments discussed above, a central premise of the argument requires some tight connection between being or existence, and goodness, greatness, or perfection. In this section, I will demonstrate that either the

145 ontological arguer relies upon CT or some close analog, or they do not. When the arguer does rely upon this, I will demonstrate this by reference to passages of their work and secondary literature. When the argument does not, I will demonstrate not only that they do not (in the same fashion, by examining primary and secondary texts), but that the arguer also fails to provide the reader with well-developed competitor theory. I will not discuss Maydole since he simply stipulates his definition connecting God to necessary existence. In this section, I will go through those arguments that do not simply stipulate a connection, and instead explore what justifications are given by those authors.

III.A Anselm

Anselm’s argument relies on CT or in a straightforward way.61 Remember, CT requires that being and goodness are the same in reference, but differ in sense. One implication of this is that insofar as there are degrees of goodness, there will be degrees of being. A further implication of this is that the opposite of goodness, evil, will be characterized as an absence or a lack of the appropriate kind of ordering or being. I do not think that Anselm explicitly discusses this connection in exactly our terms, but that it is implied in several of his written works. I will demonstrate this by referring to how

Anselm speaks about God’s goodness and existence, and how he speaks about degrees of being and goodness in other objects.

Anselm refers to the nature of God in several ways which are suggestive of (or, in rare cases explicitly about) a view that both goodness and being come in degrees, and

61 I believe, and think the reader will agree, that this approach is much more akin to the participation approach than the nature approach, as defined in Chapter 4. At this point, that is sufficient – all that is being argued is that some form of CT underlies Anselm’s argument. 146 that in God they are united. In the Proslogion, Anselm writes of God that “Therefore, you alone, of all things, exist in the truest and greatest way, for nothing else so truly exists and therefore everything else has less being,” (Anselm 2009, 7). Another translation of this passage is “You alone, then, of all things most truly exist and therefore of all things possess existence to the highest degree; for anything else does not exist as truly, and so possesses existence to a lower degree… You of all things exist to the highest degree,” (Anselm 1979, 119).

In the several different chapters of the Monologion, the operation of the convertibility thesis is frequent. For example, at the end of an argument in chapter 3 regarding how things exist with respect to other things, Anselm writes.

Accordingly, that which exists through itself exists most greatly of all. Therefore, there is some one thing which alone exists most greatly of all and most highly of all. But what exists most greatly of all and [is that] through which exists whatever is good and great and whatever is anything at all—necessarily, this is supremely good, supremely great, the highest of all existing things. Accordingly, there is something which— whether it is called a being, a substance, or a nature—is the best, the greatest, and the highest, of all existing things. (Anselm 2000b, 10, emphasis mine) In Chapter 31, he writes

… For just as that is naturally more excellent which, with respect to its natural being, more closely approximates what is most excellent, so indeed that nature exists more whose being is more like the Supreme Being… So without doubt every being exists more and is more excellent to the extent that it is more like that Being which exists supremely and is supremely excellent. (Anselm, 2000b, 48, emphasis mine) In addition to these and other various sections in Anselm’s texts that display the convertibility thesis, numerous later philosophers have interpreted Anselm in this way.

147

I believe that this point is fairly easy to establish, so I will not spend further time on supporting this thesis here.62

III.B Descartes

Descartes’ own ontological argument, as shown above, relies upon a premise linking perfection and necessary existence. Although Descartes is famously interpreted as rejecting scholasticism for modern natural philosophy, there are a number of issues where Descartes is not nearly as distinct from his scholastic predecessors than generally thought. I will show that Descartes is committed to CT or a close analog by examining the Meditations, as well as secondary sources.

In several places in the Meditations, Descartes discusses perfection as indicating greater being or reality, and imperfections as consisting in some kind of negation, absence, or non-being. For example, in discussing making mistaken judgments he tellingly writes about perfections, imperfections, and absences in being deceived and making correct judgments. I quote this passage at length;

I notice that not only a real and positive idea of God (that is, of a being supremely perfect), but also, so to speak, a certain negative idea of nothing (that is, of what is supremely lacking in every perfection) passes before me. I also notice that I have been so constituted as to be some kind of middle ground between God and nothing, or between supreme being and non-being, so that, insofar as I have been created by a supreme being, there is nothing in me by means of which I might be deceived or

62 There are numerous examples of this, but I will cite two to give the reader an idea of the establishment of this thesis. Brower 2006 argues that Anselm departs from his peers regarding certain views about goodness, but that he also holds the familiar medieval view linking goodness and being. “Anselm’s ethical theory shares much in common with that of the medieval eudaimonists. Like them, he admits a type of value that supervenes on being: things are good, he says, to the extent that they have being (M 1). Like the medieval eudaimonists, moreover, Anselm recognizes two different ways in which things can have being, and hence two corresponding types of goodness,” (Brower 2006, 229 – 230). Campbell 1980 argues for a similar interpretation of Anselm. He writes "… in the Monologion Anselm, following Augustine, erects an ascending scale— non-living, living, sentient, rational—each of which substances are greater than those before it. This scale is ranked in terms of every thing's likeness to that being which supremely is and is supremely excellent,” (Campbell 1980, 238, 239). 148

be lead into error; but insofar as I am not the supreme being, I lack quite a few things-so much so that it is not surprising that I am deceived. Thus I certainly understand that error as such is nothing real that depends upon God, but rather is only a defect… error is not a pure negation, but a privation or a lack of some knowledge that somehow ought to be in me. (Descartes 1980, 79) I think this passage is a central consideration in understanding how Descartes continues to hold some medieval or scholastic views. Similarly, in the third meditation he writes

Nor should I think that I do not perceive the infinite by means of a true idea but only through a negation of the finite, just as I perceive rest and shadows by means of a negation of motion and light. On the contrary, I clearly understand that there is more reality in an infinite substance than there is in a finite one. Thus the perception of the infinite somehow exists in me prior to the perception of the finite, that is, the perception of God exists prior to the perception of myself. Why would I know that I doubt and I desire, that is, that I lack something and that I am not wholly perfect, if there were no idea in me of a more perfect being by comparison with which I might acknowledge my defects? (Descartes 1980, 74) Again, Descartes characterizes deceit as a defect, an absence of something when he writes, at the end of the third meditation

God, I say, that same being that whose idea is in me: a being having all of those perfections that I cannot comprehend, but in some way can touch with my thought, and a being subject to no defects. From these things it is sufficiently obvious that he cannot be a deceiver. For it is manifest by the light of nature that fraud and deception depend on some defect. (Descartes 1980, 78). The correct interpretation of these passages is not altogether obvious, but I think an argument can be made fairly easily that these passages indicate a version of CT or something like it in the background of Descartes ontological argument.63

In characterizing deceit (as an example of an imperfection) as a defect, and in characterizing a supremely perfect or most perfect being as one that is “subject to no

63 This section on Descartes is the result of my own examination of the text, but my interpretation parallels and draws from the work done in Wee 2012. 149 defects” Descartes is making a connection between degrees of perfection and degrees of being, insofar as qualities and attributes are real and existent. How can a thing be defective? In the example of deceit, there might be two general kinds of causes. The first kind of defect that would cause deceit would simply be a lack of the appropriate amount of perception or intelligence – I might be deceived because of my own inability to accurately perceive the heart of the matter, or I might be deceived because I am simply not intelligent enough to realize that a deception is being perpetrated upon me.

On the other hand, the agent able to engage in deception would have an issue of will – either weakness of will that would lead them to perceive (and weakness can be characterized as a lack of a suitable degree of power) or a malicious will (and malice might be characterized as a lack of certain kinds of virtues combined with a disordered hierarchy of values). In this case, both the deceived and the deceiver suffer from some lack or imperfection.

Similarly, in discussing how his idea of finitude comes from a previously accessible idea of the infinite (and not, as initially suggested, the other way around),

Descartes explicitly characterizes certain flaws of his own as imperfections and absences. Specifically, the passage “Why would I know that I doubt and I desire, that is, that I lack something and that I am not wholly perfect, if there were no idea in me of a more perfect being by comparison with which I might acknowledge my defects?”

(Descartes 1980, 74). In this passage both doubting and desiring are immediately described as “that I lack something and that I am not wholly perfect.” Doubt is most easily characterized as an absence of knowledge and/or a certain degree of justification accessible to oneself. In order for me to doubt some belief of mine, some person’s

150 character, or that some state of affairs obtains, I must not possess a full certainty with regard to the truth of these beliefs or the propositions that represent the various states of affairs. When we think about certainty with respect to beliefs, the most obvious and straightforward way of attaining a greater degree of certainty (in either the truth or falsity of a belief) is to increase ones knowledge of surrounding details. If I doubt whether a person is trustworthy, I could in principle gather enough knowledge regarding the persons past activities and experiences, their dispositions, their track record as a friend, and so on in order to either become certain that the person is trustworthy, or certain that they are not.

Desire can also be characterized as a general kind of absence or lack. Desire only occurs in the case where one does not have an object, attain a state, or are in a situation that they wish to have, attain or be in. If an individual had that object, or were in that state or situation, they would have no desire – their desire would have been satisfied, and would no longer exist. Cecilia Wee, in discussing the meditator’s understanding of desire as a lack is to couple desire as a lack of an object in addition to the lack of the degree of power necessary to obtain that object (Wee 2012, 30).

Although this is a dialectical move for the meditator (or Descartes) to bring out that it is the idea of the infinite that is primary, I think that one need not add in omnipotence to bring out the way in which desire can be thought of as a lack or as an imperfection.

When I desire a peach, it is because I both a) have a want for it and b) do not currently have it. Insofar as desire is rooted in a kind of volitional or emotional absence, omnipotence would only change the duration of one’s desire. So perhaps God would have a desire and then immediately bring about the state of affairs that filled it, but there

151 would be a duration where God did not possess the feeling of satiety or satisfaction accompanied by a lack of desire – otherwise he would not have brought the state of affairs about in the first place. The only way for there to not be desire, at all, is for a certain mental or emotional state to be existent and ever-present – the state of being content with the way things are. Desire can only occur in the absence of this state.

Finally, the first passage quoted contains the idea of degrees of perfection correlated with degrees of being. God is described as “a being supremely perfect” and nothing is described as “what is supremely lacking in every perfection.” He also writes

“I also notice that I have been so constituted as to be some kind of middle ground between God and nothing, or between supreme being and non-being,” (Descartes 1980,

79). Here, God is not just the summit of perfection, but also the summit of being, and nothing is likewise the nadir of not only perfection, but being as well. Wee agrees, writing of this section that

She now posits a hierarchy with God, as supremely perfect being, at the top, herself as finite and lacking the perfection of God in the middle, and ‘nothingness’ (nihilum) or the complete absence of perfection at the bottom. Significantly, nihilum – utter absence – is not only utter lack of perfection: nihilum or nothingness is also an utter absence of being. Thus, the meditator re-states her hierarchy in terms of being: she herself (insofar as she is ‘lacking in countless respects’) is now in the middle between God, the highest being, and utter non-being. The meditator in this passage clearly equates perfection with being. On the metaphysics postulated in Passage C, any form of perfection is being and is a reality of some sort (whether objective or formal). Again, any imperfection is a lack, an absence of being. Thus, GOC1, which states that ‘All imperfections are absences of being’, would hold. (Wee 2012, 36-37) It is unclear whether goodness and perfection can be substituted in this passage or others, but I want to press a final point on this. Wee argues that in this passage, two general ontological claims are shown and believed by Descartes (or, alternatively, the

152 meditator); GOC 1 is that “All imperfections are absences of being” and GOC 2 that

“All forms of being are perfections,” (Wee 2012, 33). Note that these two general claims parallel CT and one of the implications of CT. GOC 1 corresponds to the inverse of CT, which is that “No beings, insofar as they are beings, are evils.” GOC 2 corresponds to one formulation of CT, that “All beings, insofar as they are beings, are good.” I hold that it is either CT or some very close analog at work in Descartes argument.

III.C Plantinga

Plantinga’s discussion of maximal greatness and great-making properties is a staple of many contemporary discussions of the ontological argument. Because of this,

I think it is surprising how little time Plantinga spends giving evidence or support for the Linkage Premise in his argument. In Plantinga, the Linkage Premise is given as follows;

(34) The property has maximal greatness entails the property has maximal excellence in every possible world. (35) Maximal excellence entails omniscience, omnipotence, and moral perfection.” (Plantinga 1974, 214) Why should we think that maximal greatness requires necessary existence?

Plantinga writes “the greatness of a being in a world W depends not merely on the qualities it has in W; what it is like in other worlds is also relevant.”

Directly after giving a version of Anselm’s argument, Plantinga writes

What could St. Anselm have meant? He takes it for granted that some beings are greater than others… Such qualities as life, consciousness, knowledge, wisdom, moral excellence, power, courage, and the like are what we might call great-making properties; the more of these properties

153

a being has and the greater degree to which it has them, the greater, all else being equal, it is.” (Plantinga 1974, 199) This is somewhat helpful, although at this point it is merely an appeal to intuitions supported by reference to some very brief cases – that a human being with courage and wisdom is greater than one without, that a human is greater than a cat because of greater degrees of intelligence and knowledge. He continues, regarding the subject of comparing and weighting of different properties (a centrally important part of any account of greatness and great-making properties)

Of course, there will be appropriate weightings; perhaps the modest degree of wisdom displayed by your average candidate for public office counts for more than the cheetah’s singular locomotive swiftness; and no doubt moral excellence outweighs power. Further, there may be cases where comparison with respect to greatness is difficult or impossible; how shall we compare a really splendid inanimate object—the Grand Teton, let us say—with a fairly undistinguished living thing—an earthworm, perhaps? Or how to compare the latter with a number? (Plantinga 1974, 199) The most important point throughout is in discussing that greatness is supposed to be contrastive. In determining a things greatness (in addition to relying on the heretofore unspecified account of great-making properties and whatever counts as a great-making property) you compare it with some other version of itself in another state of affairs.

This path leads to mostly mistakes and accidents in terms of thinking about maximal greatness.64 It is after this series of dead ends that Plantinga begins developing a conception of greatness where the greatness of a thing is not simply world indexed, but that the greatness of a thing is indexed to the degree to which it exists – the

64 The next several sections in this chapter of The Nature of Necessity explain how thinking about greatness only in this way leads to invalid or infelicitous versions of the argument. See especially pp. 201-213. 154 robustness of its existence, as expressed by how many possible worlds it exists within.

This is my terminology, as Plantinga puts it a different way. Nonetheless, it follows from what Plantinga says. Greatness of an object or being is necessarily related to not only the number of possible worlds it exists within, but also the degree to which it obtains those properties excellent for it in specific possible worlds. As he says earlier,

“The limiting degree of greatness, therefore, would be enjoyed in a given world W only by a being who had maximal excellence in W and in every other possible world as well.” The highest degree of greatness also correlates with the highest degree of existence across possible worlds – necessary existence. To summarize – greatness involves degrees of excellence of a thing, allows comparison of degrees of greatness relative to different kinds of things, and is directly tied to both the presence and the relevant qualities of a thing in other possible worlds.

This is all fine and good as far as it goes, but notice that most of what has been on offer in this discussion is a series of intuitions and stipulations – that moral excellence ranks above power, that a human being is greater than a non-sapient living animal, that a modest degree of certain great-making properties count more (in terms of greatness overall) than a high degree of other great-making properties, and that degrees of existence (captured by the number of possible worlds in which a being exists) are directly relevant to an evaluation of the greatness of an entity. No real argument has been presented, no detailed theory accounting for why greatness has a specific connection to moral traits, or degrees of existence has been given. Although I will spend more time on this in the next section, I note that some version of CT would

155 explain and undergird these claims that Plantinga makes connecting greatness to modes of existence.

III.D Gödel

Gödel’s argument is only a couple of pages of handwritten notation saved by his student, Dana Scott. As such, there is precious little information from Gödel himself about the structure of his argument, fully worked out definitions of his terms, reasons to accept specific axioms, etc. Because of the relative paucity of information given in

Gödel’s own discussion, any full account of what exactly Gödel means in his discussion of positive properties will require looking at the interpretations of secondary sources.

Therefore, although I mention Gödel’s own discussion in the beginning of this section, the majority of my textual evidence drawing out some important implications of

Gödel’s terms will be secondary sources.

As I stated before, Gödel’s own discussion of how he means to link necessary existence to goodness or perfection is quite simple. A primitive definition in Gödel’s argument is that of a positive property. Regarding a positive property, he writes

(transcribed from handwritten notes and annotated by Jordan Howard Sobel)

Positive means positive in the moral aesthetic sense (independently of the accidental structure of the world). Only then is the ax. true. It may also mean pure ‘attribution’* as opposed to ‘privation’ (or containing privation). This interpret. Simpler proof…. *i.e. the ‘disj’ normal form in terms of elem. prop. contains a member without negation. (Sobel 2003, 145) Gödel has two parts of what is meant by a positive property – it is positive in the moral/aesthetic sense, and it is positive in the sense of pure attribution as opposed to

156 privation. Even without using the terms “good” or “perfection” Gödel’s definition of a positive property should already be ringing some familiar bells in the readers head.

My own gloss on positive properties in Gödel’s sense is that they must mean something in the realm of goodness described by CT. A property’s being positive simply by being positive “in the moral/aesthetic sense” wouldn’t show that by itself – after all, people can meaningfully talk about goodness in both morality and in aesthetics without holding CT or a variant. Similarly, a property’s being positive in the sense of pure attribution as opposed to privation need not lead to any view requiring CT. After all, CT is not a commonplace in discussing qualities of temperature fluctuation, but meaningful talk about heat being a positive property and cold being one of privation (or light and dark, for example) would not be out of place. However, when we take a positive property to be both “positive in the moral/aesthetic sense” and “positive in the sense of pure attribution as opposed to privation” it is difficult to see how the end result could end up not looking something like CT. Ultimately, interpreting Gödel this way is tentative and must be – the only version of the argument that exists is a page or so of (at times hardly legible) handwritten notes. Gödel never made a more formal, extensively argued and rigorously defined version of this argument, and so interpreting this piece of work at all requires looking at Gödel’s correspondence, conversations with students, and other remarks in his diaries.

Fortunately, I am not alone in holding this interpretation of a Gӧdelian positive property. Though some authors either interpret Gödel differently or do not make any explicit mention of this one way or another (see Anderson 1990 and Look 2006), a handful of commentators have proposed that Gödel’s positive property is similar (or

157 identical) to understanding being and goodness convertibly. For example, Jordan

Howard Sobel argues this explicitly in Logic and Theism (Sobel 2003, 119). Robert

Koons also interprets Gӧdel this way in “Sobel on Gödel’s Ontological Proof,” (Koons

2006). Others have suggested, based on conversations with Gödel, that he saw his own project as either a continuation or a vindication of Leibniz’s argument.65 Since Gödel’s actual notes and the sparse notes of Dana Scott contain almost nothing as explanatory support for these definitions of positive properties, showing that several other philosophers interpret Gödel’s talk as being explicitly connected with CT suffices for the purposes to be served in this section.

III.E Malcolm

Norman Malcolm, alongside Plantinga and Maydole, does not attempt to give an entire theory that would provide justification or support for his own Linkage Premise.

Here, I explore exactly how Malcolm tries to support the relationship between necessary existence and perfection.

In Malcolm’s view, Anselm actually has two arguments for God’s existence in the Proslogion. The first relies upon the idea that existence is a perfection, but according to Malcolm (along with several other scholars), Anselm’s second ontological argument requires that necessary existence is a perfection (or is a great-making property).

Malcolm fleshes this out by thinking about the connection between contingent objects and necessary ones. When we make judgments about whether or not a given

65 For example Look, although he does not explicitly suggest that Gӧdel is committed to CT, gives reason to think that Gӧdel saw himself as continuing Leibniz’s work by referring to Wang 1987, in Look 2006. 158 object is greater than another (or more perfect than another) one of the many criteria we might use is that of dependence of the object for its continued existence. This does not exhaust our evaluative system, nor is it the most overriding factor in judgments of greatness; it is obvious that making a comparative judgment between a necessarily existing pair of scissors and a contingent moral saint is not as simple as determining which kind of thing exists necessarily. However, when we compare two copies of the same entity or object, and make one of them (in terms of the concept) a necessarily existing object or entity, Malcolm suggests that we will make the judgment that the necessarily existing copy is greater than the contingent copy. For example, he writes that

If a housewife has a set of extremely fragile dishes, then as dishes they are inferior to those of another set like them in all respects except that they are not fragile. Those of the first set are dependent for their continued existence on gentle handling; those of the second set are not. There is a definite connection in common language between the notions of dependency and inferiority, and independence and superiority. To say that something which was dependent on nothing whatever was superior to (‘greater than’) anything that was dependent in any way upon anything is quite in keeping with the everyday use of the terms ‘superior’ and ‘greater.’ (Malcolm 1960, 47) In Malcolm there is not so much a theory of great making properties as there is an examination of how our patterns of language work. In saying that the judgments about superiority and inferiority are connected to ideas of dependence and independence, Malcolm is making a point about how we understand the relationship between these concepts. All other things being equal, two exactly similar items are judged differently when we examine their degree of dependence on or independence from other factors.

159

Although there is a direct route from independent existence to necessary existence, Malcolm instead focuses on how the concepts of dependence and independence relate to the concepts of limited and unlimited entities. When we talk about an object or entity’s dependence we mean that it is limited in specified ways. If we think again of the fragile dishes, we’ll note that one of its limits is its brittleness, its fragility. For Malcolm, this is just another way to talk about dependence and independence. “It is the same thing to say that an engine's operation is dependent on as that it is limited by its fuel supply.” (Malcolm 1960, 47).

Greatness is connected to independence/being unlimited, which is (on

Malcolm’s view) one way to spell out what it would mean for God’s existence to be necessary. Critics have taken Malcolm to task about the connection between necessary existence and eternality or unlimitedness, but that is not important for our purposes

(Oppy 1995, 73). The point here is to show that there are sparse resources for a great- making theory – a few pages reflecting on linguistic usage patterns and linking together judgements of superiority to independence, independence to unlimitedness, unlimitedness to not being able to have a beginning at a point in time and not being able to have a beginning point in time to necessary existence.

III.F Leibniz

Leibniz’s views on the relationship between being and goodness (or existence and goodness) are quite complex – any attempt to talk about them with any brevity will inevitably fall short of capturing even a fraction of Leibniz’s metaphysics or the way his thought changed over time. Nonetheless, I think there is good evidence that Leibniz

160 held some version of CT or an analog, although that may only have been a part of his view on the relationship between existence and perfection.

There are two major strands in Leibniz’s body of work that suggest that Leibniz commits himself to CT. First, at several places he explicitly identifies existence with perfection. Second, he explains evil (and other imperfections) as a lack or absence of existence. Although he does not, at any point, refer to the convertibility thesis by name, these two strands in his work show that he is committed to it.

In support of the thesis that Leibniz, at several points identifies existence as a perfection, refer to his remarks about existence being a perfection in “Two Notations for

Discussion with Spinoza” (Leibniz 1970, 167 – 170), in his summer 1667 “Letter to

Arnold Eckhard” (Leibniz 1970, 177 – 181), a lemma given in 1678 from Samtliche

Schriften und Briefe (Adams 1998, p. 163), annotations to that lemma a page later in the same volume (Adams 1998, p. 164), as well as in a statement Opuscules et fragments inedits de Leibniz (Adams 1998, 166, footnote 16). From the other way, perfection is defined as the degree of reality an entity contains or evinces in “The Monadology”

(Leibniz 1970, 643 – 653), in his summer 1667 “Letter to Arnold Eckhard” (Leibniz

1970, 177 – 181), in “On the Radical Origination of Things” (Leibniz 1970, 486 – 491) and in several other places throughout his work (Blumenfeld 1994b, 393). Furthermore,

Loemker says that “That degrees of perfection and degrees of reality correspond is the common neo-Platonic assumption of Leibniz and Spinoza,” in footnote two of his translation of “Two Notations for Discussion with Spinoza.”

The other reason to suppose that Leibniz is committed to CT or some close relative is his understanding of evil as an imperfection. In the Theodicy he writes

161

20. …the formal character of evil has no efficient cause, for it consists in privation, as we shall see, namely, in that which the efficient cause does not bring about. That is why the Schoolmen are wont to call the cause of evil deficient. 21. Evil may be taken metaphysically, physically and morally. Metaphysical evil consists in mere imperfection, physical evil in suffering, and moral evil in sin. (Leibniz 1985, 136, italics mine) Donald Blumenfeld writes that Leibniz “adopts the old Scholastic view that evil is a limitation, i.e., a privation of being or perfection… this account of metaphysical evil accords with his identification of perfection with quantity of essence, or positive reality,” (Blumenfeld 1994b, 399).

Leibniz eventually developed a different view of both existence and perfection rather than those proposed (e.g. his understanding of perfection comes to be much more complicated than simply “a degree or quantity of reality” as demonstrated by

Blumenfeld 1994b, Adams 1998, Seeskin 1978, and Webb 1989). Because of this change, it could be argued that the full, mature thought of Leibniz requires no Linkage

Premise for the ontological argument. I will note two reasons that this evolution of thought in Leibniz should not count against my argument that he is committed to CT or a CT analog of some kind, drawn directly from secondary work on Leibniz. The argument here will proceed as follows – I will first show that Leibniz is committed to the truth of the claim that “existence is perfection” even when he offers other definitions of existence. Then, I will show that one can accept the claim that “perfection is variety in simplicity” since, On Leibniz’s view, this claim can be made equivalent to the claim that degrees of perfection correlate with degrees of reality or being.

First, after considering a passage where Leibniz defines existence as what would please or displease the most powerful mind, Adams argues that

162

Leibniz says that this alternative definition ‘comes to the same thing as’ the definition in terms of perfection which we have been examining… the two definitions are complementary, each encapsulating elliptically, with different omissions, the same causal account of the nature of existence. (Adams 1998, 168) For Adams, the most consistent interpretation of Leibniz’s late assertions that existence is a perfection even after it seems that he has abandoned this notion is to hold that Leibniz takes the two to be incomplete but true parts of the definition of existence.

If he is right, then we can argue that the statement “existence is a perfection” is true albeit incomplete – while it does not get at the whole of what existence is, it does get at some important truth about the nature of the relationship between existence and perfection.

Second, in order to substantiate the claim that Leibniz holds CT, we must examine whether or not his concept of perfection and its relationship to existence or reality is the same as (or close to the same as) what CT requires regarding the relationship between perfection and existence/reality. One of the definitions of perfection in Leibniz is the variety/simplicity thesis – a thing is the more perfect just in case it is a harmonious blend of simplicity and variety. In Leibniz’s own words, “God has chosen that world which is the most perfect, that is to say, which is at the same time the simplest in its hypotheses and the richest in phenomena,” (Leibniz 1970, 306).

Although this is not clearly related to the definition of perfection as consisting in the degree of reality an entity possesses, Blumenfeld hold that these two formulations equally true. For example, Blumenfeld writes

Leibniz equates a world’s perfection, or harmony, with its quantity of essence, and he thus understandably regards the best world as the one with the greatest quantity of essence. But he also thinks the world with the most harmony has the largest number of phenomenal things and the

163

largest number of monads. It follows that the most harmonious world is at once the one with the most phenomenal entities, the most monads, and the greatest quantity of essence. This brings together in a simple way a variety of Leibnizian theses: the best possible world = the most harmonious one = the one that satisfies the variety/simplicity criterion = the one with the most phenomenal individuals and the most monads = the one with the most reality or essence.” (Blumenfeld 1994b, 394) In the case of both Adams and Blumenfeld, there is a way to trace the path from viewing existence as a perfection to the path that perfection is simplicity in variety. For

Adams, the connection is that these differing views reflect the way our language and concepts are not complete – even though it is not clear, on his view, how we can make the connection between these two different views of perfection. For Blumenfeld,

Leibniz’s views on perfection, differ though they may, ultimately come to the same thing.

I would not desire to contest the idea that Leibniz’s views changed over time, or even that there are distinct formulations of the ontological argument in his work. In particular, I want to note that while Leibniz’s own metaphysics may actually be more complicated than CT, it at the very least contains CT as a proper part. If this is correct, as I’ve argued above, then at the very least, some form of Leibniz’s ontological argument required a Linkage Premise, and that the Linkage Premise is supported by CT or some similar analog.

In this section, I have demonstrated that each of these prominent authors have either a) attempted to justify the Linkage Premise with CT or some close analog, or b) attempted to try to justify the Linkage Premise without CT or anything like it. In either case (with the exception of Maydole as his argument simply stipulates this connection) all authors were at pains to show that the basic idea of the Linkage Premise, that there is

164 some strong and meaningful connection between existence or being, on the one hand, and goodness, or greatness, or perfection on the other, is justified by reflection on the nature of these concepts and their relationship with one another. In the next section, I will show that CT entails the Linkage Premise.

Section IV That the Convertibility Thesis of Being and Goodness (CT) Entails the

Linkage Premise

CT, the subject of the entire preceding chapter, is a well-worked out, detailed, and powerful theory. The conclusion of the Central Argument is that the authors we are considering ought to, at least practically, adopt CT, insofar as they are committed to the soundness of the ontological argument, and insofar as they have not already done so

(recall that some of our authors are already committed to CT or something close to it).

Each author relies on some version of the Linkage Premise for the success of their ontological argument. If CT entails the Linkage premise, then adopting CT into their conceptual framework would ensure that the Linkage Premise they give is true – it would justify the assumption of the Linkage Premise.

In previous paragraphs, I have left the Linkage Premise vague in order to ensure that it is applicable to numerous different ontological arguments. A premise in an argument is a Linkage Premise if and only if

L1. It specifies some tight connection between judgements of value such as, but not limited to, goodness, greatness, and perfection, and the status and/or strength of an entity’s being or existence. L2. It requires that the connection in L1 is, at least in part, a conceptual connection (loosely defined). L3. It requires that the connection given in L1 is correlative – the strength of the judgment of a thing’s goodness, greatness, or perfection is

165

in some way correlated to the status and/or strength of an entity’s being or existence. Given this definition, each of our previously considered arguments contain Linkage

Premises, as demonstrated in section I of this chapter.

The Convertibility Thesis holds that being and goodness are convertible. While much of the last chapter is devoted to explaining exactly what this might mean, I will give a shorter definition of CT below. CT holds that

C1. “Goodness” and “being” refer to all of the same things, or have the same extension – they are coextensive. C2. “Goodness” and “being” are necessarily coextensive – it is not a lucky accident that the classes of objects referred to by each of these terms is identical. C3. “Goodness” and “being” while not identical in sense or in intension, are nonetheless conceptually related – while they both refer to the same class of referents, it is also the case that unpacking the nature of each concept allows us to understand how goodness and being could correlate in virtue of their conceptual content. C4. Both goodness and being are matters of degree, and the degree of one is correlated to the degree of the other. If the definition given above of CT is correct, we can examine the relationship between CT and the Linkage Premise. CT is a much stronger thesis about the connection between goodness and being than the Linkage Premise. The Linkage

Premise does not require that goodness and being are coextensive, much less necessarily coextensive – just that they have some unspecified but tight relation. The

Linkage Premise does require something like a conceptual relationship between goodness and existence or being, though it leaves this underdetermined in a way that

CT does not (for a fuller statement of the conceptual relationship between the two concepts, see chapter 4). Finally, the Linkage Premise does not require that goodness

166 and being/existence come in degrees, but only that judgments made of the one are somehow correlative with judgements made about the other.

Now that we have demonstrated that CT is a stronger relation than the Linkage

Premise requires, I will show how CT entails the Linkage Premise. Remember, a premise is a Linkage Premise

L1. It specifies some tight connection between judgements of value such as, but not limited to, goodness, greatness, and perfection, and the status and/or strength of an entity’s being or existence. L2. It requires that the connection in L1 is, at least in part, a conceptual connection (loosely defined). L3. It requires that the connection given in L1 is correlative – the strength of the judgment of a thing’s goodness, greatness, or perfection is in some way correlated to the status and/or strength of an entity’s being or existence. If C1-C4 are true (that is, if CT is true), then L1 is true, for CT is a very specific and tight connection between these two concepts and judgments made using these concepts – the connection of convertibility. What does it mean to say that a connection between two concepts is tight? I am operating here with a fairly intuitive notion of

“tight conceptual connection.”66 I am confident of the judgment that convertibility is properly categorized as a tight connection.

66 On one end of the spectrum, that two concepts are identical would mean they are tightly connected – in fact, connections simply do not get much tighter, in this sense, than the identity relationship. Nearer the other end of the spectrum, we might imagine a kind of artificially drawn mereological summation or Cambridge property – though a mereological summation occurs among objects and not concepts, strictly speaking, we can imagine its conceptual analog in thinking of various Cambridge properties. Suppose that I define “trit” as “a trit is any object that either exists, or is blue, or is good, or is evil, or is red, or is shaped, or is abstract, or is coffee-flavored, or is illuminated, or is intentionally designed, or is in the region of space-time we call the Earth.” In this definition, there is a connection between goodness and existence – the connection is that they are both defining possible properties of a trit. However, this is a very loose connection, and the concepts themselves aren’t related to one another at all except through being included in tritness. Convertibility is pretty far away from this kind of connection, and in fact is much closer to identity than it or many other connections. 167

But if it is true to say “being and goodness are convertible” and that “the convertibility relationship is a tight connection between two concepts,” then it will be true to say that there is “some tight connection between judgements of value such as, but not limited to, goodness, greatness, and perfection, and the status and/or strength of an entity’s being or existence.” So CT entails L1.

CT also entails L2. While the concepts of being and goodness in CT are not, themselves, identical, it is also the case that they do not merely happen to be coextensive. It is the relationship between a things goodness and degree of goodness, and a things being or existence and its degree of being or existence that ensures that the two concepts have the same extension. Compare this to the coextension of the two terms “chordate” and “renate.” It is a matter of fact that all biological creatures on earth with kidneys (renates) also have hearts (chordates). So the terms “chordate” and

“renate” are coextensive, but not necessarily so. Things might have turned out differently in our evolutionary timeline (or they may yet, in the future, turn out differently). In the case of CT, it is the conceptual connection between being and goodness that explains their necessary coextension. Even aside from what feature of these two things ensues their joint extension, CT holds that there is a conceptual relationship between these two central concepts (again, the reader is requested to refer to Chapter 4). But if the connection between goodness and being is conceptual on CT, then L2 is true. Therefore CT entails L2.

Finally, CT entails L3. CT holds that goodness and being admit of degrees, and further that the degree of one is correlated to the degree of the other. This is just a specific way for the two concepts to be correlated with one another. Since L3 simply

168 requires that the two concepts be correlated in some way, the truth of CT suffices for the truth of L3, and therefore CT entails L3.

I have demonstrated that CT entails each proper part of the Linkage Premise.

Since CT entails both the necessary and sufficient conditions of the Linkage Premise, it entails the Linkage Premise.

Section V. The Superiority of the Convertibility Thesis

In this section, I argue that in the absence of any well-developed competitor theories, CT is the best theory to adopt in order to substantiate the truth of the Linkage

Premise. I argue this in two distinct ways. First, I argue that a holistic and detailed theory is preferred to unconnected or tenuously connected series of intuitions, observations and chains of reasoning as support for a premise. Second, I argue that even in considering CT among other possible theories, it ranks highly enough on numerous theory-decision criteria to give us a prima facie reason to accept it in order to substantiate the Linkage Premise. If this theory would be competitive with other theories, then in the absence of well-developed competitor theories we have some reason to adopt it. I also show how adopting CT rules out the AntiGod parody.

V.A The Theory Superiority Principle

The Theory Superiority Principle – A holistic, systematic theory is

always preferable to an unconnected or tenuously connected series of

intuitions, observations and chains of reasoning (ceteris paribus).

169

In this section, I give reasons to show that the Theory Superiority Principle, as given above, is plausible. If that is true, then it is reasonable to conclude that CT is preferable as a way to substantiate the Linkage Premise.67

The Theory Superiority Principle seems uncontroversial at first blush – if I have a bunch of data to explain and/or make sense of, then a holistic theory will be preferred to intuitions or observations that are only loosely connected, if at all. Philosophy is, after all, a system-builder’s discipline, so that endorsing the Theory Superiority

Principle looks like an implication of doing philosophy at all. Furthermore, if I am in dialogue with an interlocutor, I would hold that the Theory Superiority Principle also holds for each of our arguments and counterarguments. I would be more impressed with the support given to a premise by a detailed and well-worked out theory than if the premise is supported by some minimal reflection or a single intuition – and I would expect that my interlocutor would make the same judgment. One reason for adopting this principle is that it, or something like it, is already at play in the standards of argument and dialogue most of us are familiar with.

In order to support this point (aside from the evidence given explicitly below and in the readers’ own experience), I invite the reader to think about David Lewis’ philosophical system (Lewis 1986). In Lewis, metaphysical, linguistic, logical, and ethical theses are all supposed to holistically mesh with one another. Motivations for

67 Though my focus in this section is justifying this principle in philosophy, it is also the case that this principle is generally accepted in a number of other fields, including almost every branch of the hard and soft sciences. Physics, biology, geology, psychology, sociology – all of these fields presume something like the Theory Superiority Principle, and for good reason. This is not simply a theme only in science and/or philosophy. Literary theory, communications, leadership, history all have such a background principle as well. This reflection on the nature of human theorizing might lend even more support to the acceptance, in philosophy, of such a principle – if we do accept such a principle, we are in fairly good company, and our acceptance of the principle will not reflect anything unique to philosophy or the philosophical method. 170 nominalism, for possible worlds realism, for Lewis’ ethical views, for his view of causation and counterfactuals, are all tied together in one intricate and well-developed theory (or, truly speaking, set of inter-related theories). Setting aside whether or not one thinks that Lewis’ theoretical framework is true, there is much to be admired and valued in such a system simply as a detailed, organically connected way of understanding the world. Many of us hold this to be a valuable enterprise – it is much more difficult to do in the contemporary environment for numerous reasons, but doing it at all (and, in particular, doing it well) is to be lauded. It is difficult to think of any prominent, recent philosophical figure that does not have in mind holistic, systematic relationships between their various views and projects as evidenced by their written work. Of course there are other individuals whose intellectual labor tends to be narrow and disconnected from other views – scientists that specialize in niche areas are examples of this. Perhaps, in philosophy, there is also a kind of observational selection effect, as a philosopher that is prominent is likely to have, by virtue of their being prominent, published and held views on a number of philosophical areas.

There are also prominent, 20th century philosophers that changed their mind to some degree or other over time: Roderick Chisholm, Bertrand Russell, and Hilary

Putnam, as examples. However, these philosophers changed their mind due to rational pressure exerted on them by others and by their own intellectual commitments in other areas. In the end, even the philosophers we think of as having an evolving opinion are examples to be brought to bear in support of the Theory Superiority Principle.

Another reason to adopt the Theory Superiority Principle is that theories are better able to resist certain kinds of dialogical charges than mere intuitions,

171 observations, or addendums. In reflecting on the dialogical back and forth of argument, critique, response, etc., interlocutors often request justifying information. If I am to accept a premise, I want to know what reason I have for accepting the premise. What considerations, reasons, arguments, etc. can be brought to bear in support of a premise?

Similarly, when I am evaluating a critique of an argument, I am interested in asking roughly the same question – what considerations, reasons, arguments, etc. can be brought to bear in support of this criticism? This concern for justification occurs at every level of argument and counterargument, of thesis and response.

The Theory Superiority principle simply states that having a theory at one’s disposal during this dialogical process is superior, all other things being equal, to having to simply rely on an intuition, observation, or some otherwise disparate piece of reasoning.

As an example, let’s consider a compatibilist view of free will. When I ask the compatibilist about their definition of free will and the justification for accepting it, for example, they might point to how it is part of a larger theory about motivation, human psychology, and the nature of the physical world. The compatibilist might attempt to show that this usage of the term has a long history, and is interwoven with other relevant theses that all hang together just so. All of this provides more rational grounding for the definition than simply a stipulated definition, or even a mere reflection about what our intuitions are when we use such a word. Situating the definition within a broader theory of human psychology and motivation, and situating this theory in history, are both positive considerations in favor of adopting the definition. Additionally, if the definition is part of some larger theory or holistic view

172 then that will likely have other implications in other areas of the free will debate – and if this is the case, these implications (whatever they may be) will not be able to be simply shrugged off as ad hoc maneuvering on the part of the compatibilist.

As another example, let’s take neo-Aristotelian virtue theory. If the answer to

“How should I think about the structure of ethical judgments and theory?” is a focus on character and virtue, I will want some explanation for preferring this over some other view. In particular, for an Aristotelian or neo-Aristotelian view I will want to know the answer to “What counts as human flourishing?” As a proponent of the view, I could simply rely on some vague intuitions, thought experiments, or minimally connected (if at all) observations to support my thesis about the content of the phrase “human flourishing.” However, this response is rationally inferior to a neo-Aristotelian account that is rooted in a full-blown theory of human nature and function. At the very least, the theory of human nature itself might give me a principled and motivated reason to adopt some specific picture of human flourishing without opening me up to the charge of special pleading or that my definition of human flourishing is, in some way, ad hoc.

In the context of the ontological argument, an opponent might reasonably ask for justification for our Linkage Premise. Just what is it, exactly, about greatness and existence (or perfection and existence) that would explain their being linked together in such a manner as is given in the argument? Many contemporary philosophers will have short responses to this question – as we’ve seen, Maydole simply stipulates a connection between perfection and supremacy, and supremacy and necessary existence, whereas Plantinga and Malcolm both give quick, reflective musings that are supposed to support this connection, but they give us little more than that (Maydole 2009,

173

Plantinga 1974, Malcolm 1960). On the other hand, if the Theory Superiority Principle is correct then CT is a more rationally supportive ground for the Linkage Premise, and ought to be preferred if it meets the required conditions.

In examining the ontological argument from the other direction, the Theory

Superiority Principle also comes into play when evaluating critiques and responses.

Why reject existence as a first-order predicate or property? One answer is that this is because viewing existence this way is strictly inconsistent with a larger theory about existence, quantification, logic, and being. How might an opponent evaluate the concept of God if, as many believe, the proponent of the argument is simply adding on necessary existence to the concept of God in an ad hoc way, reminiscent of special pleading? Why is the Linkage Premise true for God and not for other objects (if it is not true for other objects)? An answer to these questions that relies upon a holistic, systematic theory will be given higher rational appraisal than a series of intuitions, reflections, and observations (it will also cancel out the possibility of charging the arguer or argument itself with ad hoc additions or special pleading). Of course, the opponent will want to examine the viability of the theory itself, but that is another matter entirely.

In short, the Theory Superiority Principle is to be preferred for 3 reasons. First, it is tacitly assumed in our picture of philosophy and our evaluation of philosophical projects. The entire process of philosophical back and forth and dialogue can be seen as a search for a holistic, well-developed theory within which to situate specific beliefs and views or sub-views. Although there is less great system building going on now than in the past, this valuing of theory and the role it plays in philosophical dialogue remains.

174

Second, theories are positively judged in dialogue for justifying premises or corollaries by giving substantial, principled, and systematic support. Theories are better because they are more powerful, rich, and broad than mere intuitions or reflections, and the implications of and correlations to various theories arise out of an organic unity. Third, theories are positively judged in dialogue by ruling out certain negative charges such as ad hoc reasoning or special pleading. One advantage to a view supported or implied by a theory is that a simple challenge of ad hoc reasoning or special pleading holds no weight. Whether a theory’s implications are correct or incorrect, true or false, or whether the theory as a whole is justifiable or unjustifiable, one thing that cannot be said of a theory’s implications is that it is simply a disconnected proposition meant only to respond to some criticism or challenge. Similarly, a theory gives us a principled reason to defend its implications, so charges of special pleading for a given object or connection are unmotivated.

Of course, the Theory Superiority Principle does not say simply that any theory of any kind is better than any set of intuitions, observations, and loosely connected chains of reasoning of any kind. All other things being equal, the one is to be preferred to the other. This suggests that adopting the Theory Superiority Principle will prima facie support endorsing CT, although the proponent would be required to spell out whether or not CT is itself plausible due to its other philosophical and metaphysical commitments. I have argued in the previous chapter that CT is a plausible thesis even for contemporary philosophers, and I will argue in the next section that CT ranks comparatively well on the basis of theory selection criteria.

V.B The Superiority of CT on the Basis of Theory Criteria

175

When we think about exactly what makes one theory preferable to another, a number of considerations come to mind. We might think of the depth and breadth of the theory – how deep and rich is its content, and how widely might the theory be applied? We might think of the completeness of a theory – is it the beginning of a theory but not yet fleshed out, or is it complete and detailed? We might think of simplicity – how simple, how elegant, is the theory? Finally, we might think of holistic fit – how well does the theory fit with the other things I believe and/or hold to be true?

In general, these criteria are used when deciding between two or more competing theories. In the following, I will argue that CT scores highly on each of these criteria.

When judging the depth of the theory, CT must be weighted considerably. CT is a view about the nature of metaethics, and the metaphysics of ethics, as well as a view about the nature of existence, of being. Further, it is a view that has been developed and worked out in several forms by a number of philosophers spanning hundreds of years.

A competent philosopher of history, proficient in the various required languages, would be able to produce a veritable tome on the history and different forms of CT during the

Early Patristic and Medieval period.68 This doctrine has had an influence on various philosophical and theological figures of the 19th and 20th century. There are even philosophers who are yet today working out this theory in the context of contemporary analytic philosophy. The degree of richness, detail, and depth in CT must be recognized as high (for further defense of this claim, the reader is referred to chapter 4).

When it comes to breadth, CT ranks highly on that scale, as it applies to every existent object. If a thing exists, CT has something to say about it. There is not much

68 Many scholars are even now doing this – MacDonald 1993 contains a representative sample of historical and contemporary scholarship of CT. 176 more that a theory could even cover when it comes to breadth. CT also tells us something important about certain kinds of absences, lacks, or privations, making its applicability even broader.

Another of the great virtues of CT as a theory is that it explains the connection between two realms that are often taken to be completely independent if not dichotomous – the realm of facts and the realm of values. Many of 20th century philosophers, taking on-board an insight from David Hume, have treated it as a mystery how facts and values are related.69 They thought Hume had simply proven that there is no connection at all between facts and values. The aftermath of that argument and its acceptance was that, because people often accepted it, they found the connection between facts and values mysterious, a potent problem to be solved. Much of 20th century meta-ethics and metaphysics of ethics have had trouble connecting these two realms together. CT, though, allows and even requires the connection between facts and values at a most basic level. To be is to be good, and to be good is to be. Though this idea is stunning to contemporary philosophers, it has historical roots that go back to

Aristotle and Plato in some form. Ancient and medieval philosophers would be perfectly comfortable with such a view and thus with no significant fact/value dichotomy. CT is a way of looking at reality that brings these two realms together, and this is significant because while we know facts and values are connected many of us cannot demonstrate or analyze this connection. This is just further support to show that

CT has the advantage over other theories on offer in terms of comprehensiveness.70

69 Most scholars trace this to a view represented in the very last paragraph of Book 3, Section 1, Part 1, of Hume 2000. 70 I owe the considerations expressed in this paragraph to conversations with Linda Zagzebski. 177

By virtue of having been developed over hundreds of years by many different scholars and philosophers, CT is detailed and nearer to completeness than to its beginning form. While we might wonder whether any theory is ever really complete, we do often make a distinction between a young theory and a provisional theory, as compared to a more established, detailed, rich theory.

Simplicity is a notoriously difficult criterion to make use of – there are numerous formulations of it, and different philosophers find themselves disagreeing in terms of how exactly to use this criterion (and what fits this criterion). Further, simplicity itself contains a ceteris paribus clause – simplicity is only a good criterion if the two theories being compared are otherwise equal. I cannot provide a full defense of the simplicity of CT – indeed, it requires some substantive metaphysical commitments that at least some other worldviews might rule out. That being said, there are things to be said in favor of the simplicity of CT. While it is complex in its implications, it unifies two otherwise quite distinct phenomena in goodness (moral and metaphysical) and existence. In fact, CT is a smaller subset of an older view, the Doctrine of the

Transcendentals which begins in Aristotle and is further developed by various medieval philosophers.71 While its internal workings are complicated, the Doctrine of the

Transcendentals is itself a doctrine of simplification – it reduces the categories of “one”

“true” good” and “beauty” to “being,” at least (as with CT) in terms of extensionality.

Another piece of support for CT is that it is on the same footing as moral realism, insofar as anyone that thinks goodness is a real thing will have to have a theory that is ontologically as parsimonious as the CT theorist. Another relevant consideration is that

71 See Goris and Aertsen 2013 for further information on this theory and its permutations. 178

CT is a powerful theory with a variety of implications, which is not true of other possible competing theories (and thus, making relevant the ceteris paribus condition of the simplicity criterion).

Perhaps the most challenging criterion for CT is the criterion of holism – given what most of us now think to be true about the world and about all of the relevant concepts that CT touches on (goodness, evil, existence, etc.), can CT adequately fit within this framework, or is it simply another detailed, rich theory that, as unfortunately happens with many detailed, rich theories from earlier historical epochs, has turned out to be false or unjustifiable? I have endeavored, at the end of Chapter 4, to persuade the reader that on this criterion the downfall of CT has been much overstated. I have argued that its plausibility ought to be evaluated more positively, even given our contemporary world, than many philosophers have been willing. If the arguments given at the end of Chapter 4 are successful, then CT ranks more highly on this criterion than opponents of the argument might think.

Even if there were competing theories that would explain and justify the

Linkage Premise along with a host of other phenomena, how many of them would rank as highly on each of these criteria for theory selection than CT? Leibniz’s theory, while containing CT or some close analog, also requires a number of other metaphysical, ethical, and epistemic theses that might be too far for some theorists to go – surely one need not adopt Leibniz’s entire metaphysics if that persons believes the ontological argument to be valid and sound! On the other hand, when considering the scattered, brief remarks of Plantinga and Malcolm as theories, which we might simply for the sake of comparing them here, neither of these will rank as highly on breadth, depth,

179 completeness. They may or may not count as simpler, depending on what we mean by simplicity and how we judge different things to be simple. Finally, each of these views might be found to be objectionable to our contemporary world-view insofar as they require some substantial ethical and metaphysical commitments that we are hesitant to adopt without further.

Of course, simply calling the meager and chaotic reflections on the relationship between greatness and existence in Plantinga or Malcolm theories is to badly stretch the word.72 By definition, neither of these will count as theories and so will not be proper objects of comparison to CT. However, I am arguing that even if they were so counted,

CT would remain superior overall as a theory in support of the Linkage Premise and on its own merits.

As a final consideration in favor of CT as compared with other theories as supporting the Linkage Premise, note how adoption of this background theory would affect other arguments, critiques, or concepts surrounding the ontological argument. CT completely rules out the possibility of Antigod Parodies of the argument – if simply existing requires any amount of goodness, then a perfectly evil being will simply be one that completely fails to exist. If a perfectly good being is the apex of being on this view, then a perfectly evil being would not be, and in fact would be nothing. For it to exist at all it would be good to some degree, and thus not perfectly evil. Other parodies

72 I do not take this to be an indictment of those philosophers, just that they believed that this connection between existence and greatness (or maximality or perfection) is so commonly held at an implicit level that a short amount of reflection and intuition pumping would suffice to persuade their readers of the rational acceptability of considerations in support of their Linkage Premise. However, since I believe this to be a lynchpin of the argument as a whole, and since I believe that their own considerations are not sufficient to ground this thesis, I have chosen to use the language given above as an accurate, if not wholly flattering, view of their accounts of being and goodness and their relation. 180 might be more effective against an ontological argument with the backing of CT, but they have other assorted problems as explained in chapter three.

CT may also undercut the longstanding critique of the ontological argument that existence is not a perfection or a property. Existence is a perfection or great-making property in a straightforward way if one accepts CT.73 Far from being univocally radical in this way, CT is consistent with a number of theories of existence, even if it rules out or is inconsistent with one of the major, mainline theories of existence. CT may also, if accepted, lend some support to the possibility premise of the argument.

While metaphysical modal judgments do not depend on our given set of concepts and ideas, it is a psychological fact about human beings that adoption of differing worldviews will lead to differing appraisals of claims and statements. Were we inclined to accept CT prior to appraisal of the argument, I believe that we would find the possibility premise much less objectionable. Finally, CT gives some credence to the idea of God as the ens realissimum, and as the apex of being. This is not simply a boldly stated assertion or an ad hoc modification of the concept, but is the natural implication of theory of the relationship between reality and goodness.74

In conclusion, I have given a three-pronged argument in this section that CT is to be preferred in the absence of competitor theories. First, I have shown that philosophers ought to accept the Theory Superiority Principle. If this principle is accepted, it goes a long way in supporting the view that CT ought to be adopted instead of considerations given by many contemporary proponents of the ontological argument.

73 On at least some interpretations. Alexander 2012 argues that CT does not require treating existence or goodness as properties of objects. 74 Alexander 2012 suggests that his attributive account of goodness, which seems to entail CT, provides a way to give further rational support to Aquinas’ Fourth Way. 181

Second, I have shown that even if we were comparing CT to the other available theories

(which is, again, to badly stretch the word) CT would have a higher aggregate ranking on various theory selection criteria than those competitor theories. Third, I have shown that simply by itself, CT has certain advantages over other theories, not only but especially in the case of the ontological argument. If this is true, it gives us reason to adopt CT in order to substantiate the Linkage Premise in ontological arguments. If my arguments in this chapter are correct, then CT ought to be adopted by proponents of the ontological argument.

Section VI. Conclusion

VI.A Chapter Conclusion

In this chapter, I have shown that various representative versions of the ontological argument rely upon a Linkage Premise of some kind. In section II, I have demonstrated that in many cases, these arguments require CT or something very similar to it in order to provide justification and support for the Linkage Premise. Those versions of the argument that do not explicitly rely on CT either stipulate the truth of the

Linkage Premise, or they justify the Linkage Premise without any well-developed competitor theory to CT. In section III, I have shown that CT entails the Linkage

Premise. Finally, in section IV, I gave a two-pronged argument for adopting CT in the absence of well-developed competitor theories. If the foregoing sections are correct, if the arguments, reasons, and considerations given are valid and sound, then proponents of any version of the ontological argument that both a) contains a Linkage Premise, and b) that does not include a well-developed competitor theory, are practically committed to CT. To put this another way – if the proponent of the argument is interested in

182 making their version of the ontological argument as strong as possible, as justified and well-supported as possible, and if their version of the ontological argument meets those two conditions given above, they will have little choice other than to affirm CT.

Of course, this should not worry the proponent of the argument. I have shown that CT is a plausible account of the metaphysical relationship between being and goodness. If this is correct, then proponents of the argument have added a powerful and distinguished tool to their philosophical toolkit. On the other hand, this should give pause to critics of the ontological argument. If the argument contained in chapter 2 about the weakness of one major and traditional critique of the ontological argument

(that existence is not a predicate) is correct, then critics will have fewer places to attack than previously thought. If the argument contained in chapter 4 is correct, then the strongest kind of objection to ontological arguments is a kind of parody objection.

Since CT rules out this objection, accepting CT would rule out the strongest objection to the ontological argument. If this is true, then critics will have to acknowledge CT as part of the conceptual background of the ontological argument. They must wrestle with this theory of the metaphysical relationship between two of our most important ethical and metaphysical concepts. All of this is a much more complex, in-depth task than most critics of the argument have considered it to be.

VI.B Dissertation Conclusion

This dissertation has spanned a number of significant topics in metaphysics and philosophy of religion. In the first chapter, I gave a brief overview of those arguments I discuss in this chapter, as well as noted other contemporary versions of the ontological argument. In the second chapter, I examined a number of existence concepts, and noted

183 which do and do not seem to be required and/or ruled out for ontological arguments to be successful. I believe the second chapter contains the required backgrounding of certain common ways of objecting to the ontological argument – indeed, in discussing the ontological argument with any professional philosopher who does not specialize in philosophy of religion, this is the most common form of criticism. It is often thought to be so persuasive that after it is lodged, little remains to be said. In the third chapter, I argue that parody style arguments are actually the most powerful objection to the ontological argument, and I develop an account of the AntiGod objection that I argue is the most persuasive form of parody objection available. In the fourth chapter, I examine and develop the nature approach of the Convertibility Thesis of Being and

Goodness (referred to throughout this chapter as CT), in both historical and contemporary work. This chapter also includes a defense of the initial plausibility of

CT in contemporary philosophical work. In this final chapter, I have argued that those typical ontological arguments I consider in earlier chapter are either directly or practically committed to the truth of CT, as well as arguing that CT is superior to other theories on offer

If the foregoing research and series of arguments are correct, then we may conclude a number of interesting things. First, if my arguments in the third chapter are correct, the most persuasive objection to the ontological argument is some form of parody objection, especially and in particular the AntiGod or Devil-like-being objection. This is certainly not the consensus in the field at this time – specialists often think that parody objections are easily refuted and betray a misunderstanding of the structure of the ontological arguments. It seems that in some form, Gaunilo’s original

184 intuition and objection lives on to be a gadfly at Anselm’s argument. Extending this intuition to the form of the AntiGod parody (or other devil parodies) is not entirely new, but its force has long gone under acknowledged and underappreciated. Despite

Nagasawa’s arguments, parody objections are not destined to fail – if I am correct, they are the most forceful and most persuasive kind of objection to ontological arguments. If this is true, then CT has, as previously argued, another significant advantage – it completely rules out these kind of parody objections.

Second, for CT being a central and important feature of a number of important ontological arguments, the contemporary philosophical world has been almost entirely silent about this. It has not, for the most part, even been noticed as a significant feature of a number of ontological arguments, much less argued against, analyzed, or incorporated into contemporary research into the argument. Even contemporary philosophers who give their own versions of the argument seem to think it is entirely unimportant how or to what degree the Linkage Premise so often included in ontological arguments is supported by anything other than a brief series of reflections or outright stipulation. Part of this lack of attention is likely simply a matter of the more widespread lack of attention paid to medieval philosophy in the contemporary philosophical world. Unlike the works of Plato and Descartes, whose concepts and terminology every professional philosopher is steeped in since their very first philosophy class, the concepts, terminology, and arguments of medieval philosophers are absent or extremely rare in most courses of contemporary philosophical education.

This might explain why we often fail to notice the medieval influences in Leibniz or

185

Descartes, or why concepts and theories that were mainstays of swaths of medieval philosophy are so foreign to the contemporary philosophical world.

However, if this dissertation is correct, it will no longer be sufficient to ignore

CT, especially and in particular in discussions of the ontological argument. If my foregoing arguments are correct, CT is lurking behind every ontological argument that includes a Linkage Premise and that does not contain a competitor theory to CT.

Contemporary philosophers of religion (including atheologians) will be required to be explicit about their background metaphysics in a more thorough way than they previously have. They will be required to grapple with the powerful and taxing

Convertibility Thesis of Being and Goodness. This will not be an easy task on either side – simply understanding CT and its odd notions about goodness, perfection, being, and existence requires a great degree of careful study and reflection. No one, to my knowledge, has made this claim about ontological argument before now, so if I am my arguments are correct, this will require a shift in the contemporary field.

Of course, the more significant take-away is that if CT is minimally plausible, it may be worth researching and incorporating into areas far beyond the ontological argument. It would be the project of another work entirely to attempt to justify CT as a theory consistent with the contemporary philosophical world. All that I have tried to do in this dissertation is to show that instead of quickly writing off this metaethical and metaphysical theory with a passing reference to pain or the privation theory of evil, much more critical engagement will be required. If CT were accepted as a legitimate metaethical and metaphysical theory, there would be implications for general ethical theory, theories of existence, theories of the good, theories of function, etc.

186

Finally, just as the CT of being and goodness has proven to be important and useful in this contemporary philosophical debate (in addition to its wide-reaching implications in metaethics and basic metaphysics), other concepts and theories from medieval philosophy may shed light on other areas of contemporary philosophy. In particular, it may be a useful exercise to see how a defense of CTs of other transcendental categories might inform contemporary philosophy. A CT of being and oneness would prove provocative in basic metaphysics as regards the debate between pluralism and monism. A CT of being and truth might require revisions or alterations in logic, philosophy of language, and metaphysics. A CT of being and beauty might alter contemporary aesthetics and the metaphysics aestheticians are committed to.

187

References

Adams, Robert. 1998, Leibniz: Determinist, Theist, Idealist, Oxford: Oxford University Press.

Aertsen, Jan. 1985, “The Convertibility of Being and Good in St. Thomas Aquinas”, New Scholasticism 59: 449 – 470.

____. 1991, “Good as Transcendental and the Transcendence of the Good”, in Being and Goodness, S. MacDonald (ed., trans.), Ithaca: Cornell University Press, pp. 56 – 73.

____. 1996, Medieval Philosophy and the Transcendentals: The Case of Thomas Aquinas, Leiden: Brill.

____. 1999, “Thomas Aquinas on the Good: The Relation Between Metaphysics and Ethics”, in Aquinas’ Moral Theory: Essays in Honor of Norman Kretzmann, S. MacDonald and E. Stump (ed.s), Ithaca: Cornell University Press, pp. 235 – 253.

Alexander, David. 2012, Goodness, God, and Evil, New York: Continuum International Publishing Group.

Anderson, C. Anthony. 1990, “Some Emendations on Gödel’s Ontological Proof”, Faith and Philosophy, 7: 291–303.

Anglin, Bill and Stewart Goetz. 1982, “Evil is Privation”, International Journal for Philosophy of Religion, 13: 3 – 12.

Anselm. 1979, Proslogion, M.J. Charlesworth (trans.), Notre Dame: University of Notre Dame Press.

____. 2000a, Proslogion in The Complete Philosophical and Theological Treatises of Anselm of Canterbury, Jasper Hopkins and Herbert Richardson (trans.), Minneapolis: The Arthur J. Banning Press.

____. 2000b, Monologion in The Complete Philosophical and Theological Treatises of Anselm of Canterbury, Jasper Hopkins and Herbert Richardson (trans.), Minneapolis: The Arthur J. Banning Press.

____. 2007, Basic Writings, Thomas Williams (ed., trans.), Indianapolis; Hackett.

____. 2009, Proslogion, in The Many Faced Argument, John Hick and Arthur McGill (ed.s and trans.), Eugene: Wipf and Stock, 3 – 8.

Aquinas. 1920, Summa Theologiae, K. Knight (ed.), available online at http://www.newadvent.org/summa/.

188

____. 1955, Summa Contra Gentiles, J. Kenny (ed.), New York: Hanover House, available online at http://dhspriory.org/thomas/ContraGentiles.htm.

____. 2001, De Malo, Brian Davies (ed.), Richard Regan (trans.), Oxford: Oxford University Press.

Aristotle. 1984, The Complete Works of Aristotle, J. Barnes (ed.), Princeton: Princeton University Press.

Augustine. 1887, Nicene and Post-Nicene Fathers, First Series, M. Dods, P. Schaff, and K. Knight (trans. and ed.s), Buffalo: Christian Literature Publishing. Available online from http://www.newadvent.org.

____. 1950, Enchiridion, Bernard Peebles (trans.), New York: Fathers of the Church.

Aydede, Murat. 2005, Pain: New Essays on Its Nature and the Methodology of Its Study, Cambridge: MIT Press.

Baker, Samuel. 2016, “The Metaphysics of Goodness in the Ethics of Aristotle,” Philosophical Studies: 1 – 18.

Blumenfeld, Donald. 1994a, “Leibniz’s Ontological and Cosmological Arguments”, ”, in The Cambridge Companion to Leibniz, N. Jolley (ed.), Cambridge: Cambridge University Press, 353 – 381.

____. 1994b, “Perfection and Happiness in the Best Possible World”, in The Cambridge Companion to Leibniz, N. Jolley (ed.), Cambridge: Cambridge University Press, 382 – 410.

Boethius. 1991, _De Hebdomadibus_, in Being and Goodness, S. MacDonald (ed., trans.), Ithaca: Cornell University Press, 299—304.

Brecher, Robert. 1974, “’Greatness’ in Anselm’s Ontological Argument”, The Philosophical Quarterly, 95: 97 – 105.

Brower, Jeffrey. 2006, “Anselm on Ethics” in The Cambridge Companion to Anselm, B. Davies and B. Leftow (ed.s), Cambridge: Cambridge University Press: 222 – 256.

Buss, David. 2001, “Human Nature and Culture: An Evolutionary Psychology Perspective”, Journal of Personality, 69: 955 – 978.

Calder, Todd. 2007, “Is the Privation Theory of Evil Dead?”, American Philosophical Quarterly, 44: 371 – 381.

Campbell, Richard. 1980, “Anselm’s Background Metaphysics”, Scottish Journal of Theology, 33: 317 – 343.

189

Chalmers, David. 1999. “Materialism and the Metaphysics of Modality”, Philosophy and Phenomenological Research, 59: 473 – 496.

____. 2002, “Does Conceivability Entail Possibility?”, in Conceivability and Possibility, T. Gendler and J. Hawthorne (ed.s), Oxford: Oxford University Press, pp. 145 – 200.

____. 2006, “Two Dimensional Semantics”, in Oxford Handbook of the Philosophy of Language, E. Lepore and B. Smith (ed.s), Oxford: Oxford University Press, 574 – 606.

Chambers, Timothy. 2000, “On Behalf of the Devil: A Parody of St. Anselm Revisited”, Proceedings of the Aristotelian Society (New Series), 100: 93–113.

Clark, Austen. 2005, “Painfulness is Not a Quale”, in Pain: New Essays on Its Nature and the Methodology of Its Study, M. Aydede (ed.), Cambridge: MIT Press, 177 – 198.

Cock, Albert. 1918, “The Ontological Argument for the Existence of God”, Proceedings of the Aristotelian Society (New Series) 18: 363 – 384.

Davidson, Donald. 1987, “Knowing One’s Own Mind”, Proceedings and Addresses of the American Philosophical Association 60: 441-458…

Davis, Douglas. 1989, “The Privation Account of Evil: H.J. McCloskey and Francisco Suarez” in The Metaphysics of Substance, Daniel Dahlstrom (ed.), Washington, D.C.: American Catholic Philosophical Association.

Descartes, Rene. 1980, Discourse on Method and Meditations on First Philosophy, Donald Cress (trans.), Indianapolis: Hackett Publishing.

____. 2008, Meditations on First Philosophy with Selections from the Objections and Replies, Michael Moriarty (trans.), Oxford: Oxford University Press.

Devine, Philip. 1975, “The Perfect Island, the Devil, and Existent Unicorns”, American Philosophical Quarterly, 12: pp 255—60.

Dombrowski, Daniel. 2006, Rethinking the Ontological Argument: A Neoclassical Theistic Response, Cambridge: Cambridge University Press.

Elders, Leo. 1993, The Metaphysics of Being of St. Thomas Aquinas, New York: Brill.

Everitt, Nicholas. 2004, The Non-Existence of God, London: Blackwell.

Foot, Philippa. 2001, Natural Goodness, Oxford: Oxford University Press.

190

Frege, Gottlob. 1960, The Foundations of Arithmetic, J.L. Austin (trans.), New York: Harper.

____. 1980a, “Concept and Object”, Translations from the Philosophical Writings of Gottlob Frege, P. Geach and M. Black (ed. and trans.), Oxford: Blackwell.

____. 1980b, “Function and Concept”, Translations from the Philosophical Writings of Gottlob Frege, P. Geach and M. Black (ed. and trans.), Oxford: Blackwell.

Gale, Richard. 1993, On the Nature and Existence of God, New York: Cambridge University Press.

Geach, Peter. 1955, “Form and Existence”, Proceedings of the Aristotelian Society (New Series), 55: 251-272.

____. 1956, “Good and Evil”, Analysis, 17: 33 – 42.

____. 1968, “What Actually Exists”, Proceedings of the Aristotelian Society (Supplementary Volumes), 42: 7-16.

Gendler, Tamar Szabo and John Hawthorne. 2002a, Conceivability and Possibility, Oxford: Oxford University Press.

____. 2002b, “Introduction: Conceivability and Possibility” in Conceivability and Possibility, T. Gendler and J. Hawthorne (ed.s), Oxford: Oxford University Press, pp. 1- 71.

Goodman, Nelson. 1973, Fact, Fiction, and Forecast, Cambridge: Harvard University Press.

Goris, Wouter and Jan Aertsen. 2013, “Medieval Theories of Transcendentals” in Stanford Encyclopedia of Philosophy, available at https://plato.stanford.edu/entries/transcendentals-medieval/.

Gracia, Jorge J.E. 1991, “Evil and the Transcendentality of Goodness: Suarez’s Solution to the Problem of Positive Evils”, in Being and Goodness, S. MacDonald (ed., trans.), Ithaca: Cornell University Press, pp. 151—178.

Grant, W. Matthews. 2015, “The Privation Account of Moral Evil: A Defense”, International Philosophical Quarterly, 55: 271 – 286.

Gustafson, Don. 2005, “Categorizing Pain”, in Pain: New Essays on Its Nature and the Methodology of Its Study, M. Aydede (ed.), Cambridge: MIT Press, 219 – 242.

Haight, David and Marjorie Haight. 1970, “An Ontological Argument for the Devil”, The Monist, 54: 218 – 220.

191

Haight, David. 1974, “Devils”, International Journal for Philosophy of Religion, 5: 152 – 156.

____. 1994, “What Have We Learned From Ontological Devil-arguments?”, International Philosophical Quarterly, 34: 301 – 306.

Harrelson, Kevin. 2009, The Ontological Argument from Descartes to Hegel, New York: Humanity Books.

Hartshorne, Charles. 1965, Anselm's Discovery: A Re-Examination of the Ontological Proof for God's Existence, La Salle: Open Court.

Hazen, Allen. 1999, “On Gödel’s Ontological Proof”, Australasian Journal of Philosophy, 76: 361–377.

Henle, Paul. 1961, “Use of the Ontological Argument”, The Philosophical Review, 70: 102 – 109.

Hick, John, and Arthur McGill. 2009, The Many-Faced Argument, Eugene: Wipf and Stock Publishers.

Hume, David. 2000, A Treatise on Human Nature, Oxford: Oxford University Press.

Hursthouse, Rosalind. 1999, On Virtue Ethics, Oxford: Oxford University Press.

Johnson, Daniel. 2009, “The Sense of Deity and Begging the Question with Cosmological and Ontological Arguments”, Faith and Philosophy 26: pp. 87—94.

Johnson, Oliver. 1965, “God and St. Anselm”, Journal of Religion, 45: 326 – 334.

Kane, G. Stanley. 1980, “Evil and Privation”, International Journal for Philosophy of Religion, 11: 43 – 58.

Kane, Robert. 1984, “The Modal Ontological Argument”, Mind, 93: 336 – 350.

Kenny, Anthony. 2002, Aquinas on Being, Oxford: Oxford University Press.

Klein, Colin. 2007, “An Imperative Theory of Pain”, Journal of Philosophy, 104: 517 – 532.

____. 2015, “What Pain Asymbolia Really Shows”, Mind, 124: 493 – 516.

Koons, Robert. 2006, “Sobel on Gӧdel’s Ontological Proof”, Philosophia Christi, 8: 235 – 248.

192

Larrimore, Mark. 2011, “Evil as Privation: Seeing Darkness, Hearing Silence” in Deliver Us From Evil, M.D. Eckel and B.L. Herling (ed.s), London: Continuum, pp. 149 – 165.

Leftow, Brian. 2005, “The Ontological Argument”, in Oxford Handbook of Philosophy of Religion, W. Wainwright (ed.), Oxford: Oxford University Press, pp. 80–115.

Leibniz, Gottfried Wilhelm. 1970, Philosophical Papers and Letters, Leroy Loemker (trans.), Dordrecht: D. Reidel.

____. 1985, Theodicy: Essays on the Goodness of God, the Freedom of Man, and the Origin of Evil, E.M. Huggard (trans.), La Salle: Open Court.

____. 1992, De Summa Rerum, G.H.R. Parkinson (trans.), New Haven: Yale University Press.

Lewis, David. 1970, “Anselm and Actuality”, Noûs, 4: 175–88.

____. 1986, On the Plurality of Worlds, Oxford: Blackwell Publishing.

Look, Brandon. 2006, “Some Remarks on the Ontological Arguments of Leibniz and Goedel” in Einheit in der Vielheit: Akten des VIII, H. Breger (ed.), Hannover: Hartmann, 510 – 517.

____. 2013, “Gottfried Wilhelm Leibniz”, in Stanford Encyclopedia of Philosophy, available at https://plato.stanford.edu/entries/leibniz/.

Lowe, E.J. 2007, “The Ontological Argument”, in The Routledge Companion to Philosophy of Religion, P. Copan and C. Meister (eds.), London: Routledge.

MacDonald, Scott. 1989, “Augustine’s Christian-Platonist Account of Goodness”, New Scholasticism, 63: 485—509.

____. 1991a, Being and Goodness, Ithaca: Cornell University Press.

____. 1991b, “The Relation Between Being and Goodness” in Being and Goodness, S. MacDonald (ed.), Ithaca: Cornell UP, 1 – 28.

____. 1991c, “The Metaphysics of Goodness and the Doctrine of the Transcendentals” in Being and Goodness, S. MacDonald (ed.), Ithaca: Cornell UP, 31—55.

MacDonald, Scott and Eleonore Stump. 1999, Aquinas’ Moral Theory: Essays in Honor of Norman Kretzmann, Ithaca: Cornell University Press.

Malcolm, Norman. 1960, “Anselm's Ontological Arguments”, Philosophical Review, 69: pp. 41–62.

193

Mann, William. 1976, “The Perfect Island”, Mind, 85: 417 – 421.

Matthews, Gareth and Lynn Baker. 2010a, “The Ontological Argument Simplified”, Analysis, 70: pp. 210–212.

____. 2010b, “Anselm’s Argument Reconsidered”, Review of Metaphysics 64: pp. 31— 54.

Maydole, Robert. 2003, “The Modal Perfection Argument for the Existence of a Supreme Being”, Philo 6: 299-313.

____. 2009, “The Ontological Argument”, in The Blackwell Companion to Natural Theology, W. Craig and J. Moreland (ed.), Oxford: Blackwell, 553–592.

McInerny, Ralph. 1991, “Saint Thomas on De Hebdomadibus”, in Being and Goodness, S. MacDonald (ed.), Ithaca: Cornell UP, 74 – 97.

Midgely, Mary. 1984, Wickedness: A Philosophical Essay, London: Routledge & Kegan Paul.

Millican, Peter. 1989, “The Devil’s Advocate”, Cogito, 3: 193—207.

____. 2004, “The One Fatal Flaw in Anselm's Argument”, Mind, 113: 437–76.

____. 2007, “Ontological Arguments and the Superiority of Existence”, Mind, 116: 417 – 421.

Moser, Paul. 1987, A Priori Knowledge, Oxford: Oxford University Press.

Nagasawa, Yujin. 2010, “The Ontological Argument and the Devil”, Philosophical Quarterly, 60: pp. 72—91.

Nussbaum, Martha. 1992, “Human Functioning and Social Justice: In Defense of Aristotelian Essentialism”, Political Theory, 20: 202 – 246.

____. 2000, Women and Human Development: The Capabilities Approach, Cambridge: Cambridge University Press.

Oppenheimer, Paul and Edward Zalta. 2011, “A Computationally-Discovered Simplification of the Ontological Argument”, Australasian Journal of Philosophy, 89: 333–349.

Oppy, Graham. 1995, Ontological Arguments and Belief in God, New York: Cambridge University Press.

194

____. 2011, “Objection to a Simplified Ontological Argument”, Analysis, 71: pp. 105— 106.

Parsons, Terence. 1980, Nonexistent Objects, New Haven: Yale University Press.

Plantinga, Alvin. 1965, The Ontological Argument, Garden City: Anchor Books. ____. 1974, The Nature of Necessity, Oxford: Oxford University Press.

____. 1977, God, Freedom, and Evil, Grand Rapids: W.B. Eerdman’s Pub. Co.

Plato. 1997, Complete Works, J.M. Cooper and D.S. Hutchinson (ed.s), Indianapolis: Hackett.

Pruss, Alexander. 2009, “A Gӧdelian Ontological Argument Improved”, Religious Studies, 45: 347–353.

____. 2010,“The Ontological Argument and the Motivational Centres of Lives”, Religious Studies, 46: 233–249.

Pseudo-Dionysius. 1987a, Pseudo-Dionysius: The Complete Works, C. Luibheid and P. Rorem (trans.), New York: Paulist Press.

____. 1987b, The Divine Names in Pseudo-Dionysius: The Complete Works, C. Luibheid and P. Rorem (trans.), New York: Paulist Press, pp. 47 – 132.

Quine, Willard van Orman. 1948, “On What There Is”, Review of Metaphysics, 2: 21- 38.

Reeve, C.D.C. 2003, “Plato’s Metaphysics of Morals”, Oxford Studies in Ancient Philosophy 25: 39 – 58.

Richman, Robert. 1958, “The Ontological Proof of the Devil”, Philosophical Studies, 9: 63 – 64.

____. 1960, “The Devil and Dr. Waldman”, Philosophical Studies, 11: 78 – 80.

Rowe, William. 1976, “The Ontological Argument and Question-Begging”, International Journal for the Philosophy of Religion, 7: 425 – 432.

Russell, Bertrand. 1905, “The Existential Import of Propositions,” Mind, 55: 398-401.

____. 2010, The Philosophy of Logical Atomism, London: Routledge.

Salmon, Nathan. 1989, “The Logic of What Might Have Been,” Philosophical Review, 98: 3 – 34.

195

Schindler, David. 2005, “What’s the Difference? On the Metaphysics of Participation in a Christian Context”, The Saint Anselm Journal, 3.1: 1 – 27.

Seeskin, Kenneth. 1978, “Is Existence a Perfection? A Case Study in the Philosophy of Leibniz”, Idealistic Studies, 8: 124 – 135.

Shields, Chris. 2011, “Surpassing in Dignity and Power: The Metaphysics of Goodness in Plato’s Republic”, in Socratic, Platonic, and Aristotelian Studies: Essays in Honor of Gerasimos Santas, Georgios Anagnostopoulos (ed.), London: Springer, pp. 281 – 296.

Sobel, Jordan. 1987, “Gödel’s Ontological Proof” in On Being and Saying: Essays for Richard Cartwright J.J. Thomson (ed.), London: MIT Press, 241-261.

____. 2003, Logic and Theism, New York: Cambridge University Press.

Stenmark, Mikael. 2009, “Three Theories of Human Nature”, Zygon, 44: 894 – 920.

____. 2012a, “Theories of Human Nature: Key Issues”, Philosophy Compass, 8: 543 – 558.

____. 2012b, “Is There a Human Nature?”, Zygon, 47: 890 – 902.

Stump, Eleonore, and Norman Kretzmann. 1991, “Being and Goodness” in Being and Goodness, S. MacDonald (ed.), Ithaca: Cornell University Press, pp. 98—128.

Suarez. 1989, The Metaphysics of Good and Evil according to Suarez: Disputations X and XI, Jorge J.E. Gracia (trans.) and Douglas Davis (trans.), Analytica Series, Munich and Vienna: Philosophia Verlag.

Swenson, Adam. 2009, “Privation Theories of Pain”, International Journal for Philosophy of Religion, 66: 139 – 154.

Tooley, Michael. 1981, “Plantinga's Defence of the Ontological Argument”, Mind, 90: 422–7.

Tye, Michael. 2005, “Another Look at Representationalism About Pain”, in Pain: New Essays on Its Nature and the Methodology of Its Study, M. Aydede (ed.), Cambridge: MIT Press, 99 – 120.

Vaidya, Anand. 2015, “The Epistemology of Modality” in Stanford Encyclopedia of Philosophy, available at https://plato.stanford.edu/entries/modality-epistemology/. van Inwagen, Peter. 1977, “Ontological Arguments”, Noûs, 11: 375–395.

____. 1998, “Modal Epistemology”, Philosophical Studies, 92: 67-84.

196

____. 2002, Metaphysics, 2nd edition, Boulder: Westview Press.

____. 2009, “Some Remarks on the Modal Ontological Argument”, Philo, 12: pp. 217—227.

____. 2014, “Being, Existence, and Ontological Commitment” in Existence: Essays in Ontology, New York: Cambridge University Press, pp. 50 – 86.

Wang, Hao. 1987, Reflections on Kurt Gӧdel, Cambridge: MIT Press.

Webb, Mark. 1989, “Natural Theology and the Concept of Perfection in Descartes, Spinoza and Leibniz”, Religious Studies, 25: 459 – 475.

Wee, Cecilia. 2012, “Descartes Ontological Proof of God’s Existence”, British Journal for the History of Philosophy, 20: 23 – 40.

Wilbanks, Jan. 1973, “Some (Logical) Trouble for St. Anselm”, New Scholasticism, 47: 361 – 365.

Yablo, Stephen. 1993, “Is Conceivability a Guide to Possibility?”, Philosophy and Phenomenological Research, 53: 1 – 42.

____. 1999, “Concepts and Consciousness”, Philosophy and Phenomenological Research, 59: 455 – 463.

____. 2002, “Coulda, Woulda, Shoulda”, in Conceivability and Possibility, T. Gendler and J. Hawthorne (ed.s), Oxford: Oxford University Press, 441 – 492.

197